7

ECONOMIC AND NATIONAL SECURITY RISKS PERTAINING TO THE BIOECONOMY

As described in Chapter 3, the bioeconomy represents an important and growing share of the U.S. economy. The committee believes the bioeconomy’s importance will continue to grow as biotechnology makes greater inroads in pharmaceutical production and the delivery of health care, in agriculture, in the generation of energy, in the manufacture of specialty chemicals and materials, and in the production of other goods and services, particularly as biological production processes displace conventional chemical processes. The bioeconomy is also important to national defense—not only in the narrow sense of countering biological weapons but also for a broader range of defense needs (DiEuliis, 2018), including military medicine (NRC, 2004); sensors, electronics, computing, materials, logistics, and soldier health and performance (Armstrong et al., 2010; NRC, 2001; Tucker, 2019); and energy (NRC, 2012). In these defense-related applications, biotechnology is a thoroughly dual-use¹

___________________

¹ The term “dual-use” has two related but distinct meanings. With respect to export controls, it refers to items produced for commercial markets that can also be used in military systems, and that therefore are subject to national security export controls. With respect to scientific research, the term also refers to legitimate scientific developments that can be misused for harm. Biotechnology is dual-use in both respects, recognizing that there are military uses that do not involve the development and production of biological weapons, which is banned by the international Biological Weapons Convention.

technology, meaning that the same science and technology base underlies both military and commercial applications, making it difficult to distinguish the economic security and national security aspects of the technology. This ambiguity is exacerbated by the continued emergence and evolution of new biotechnologies whose ultimate applications and significance for the economy or for national security may not yet be clear. Even if the economic or national security impact of specific technologies could be determined unambiguously, these two areas are interrelated in that a country’s economic vitality affects its ability to support its national defense and other national needs. Moreover, a nation that is unable to support an economically vital industrial sector is potentially vulnerable to coercion or monopoly pricing from foreign suppliers. Given this blurring of economic and national security concerns, much of this chapter’s discussion does not differentiate economic from national security risks.

The first section of the chapter addresses potential harms to the health and competitiveness of the U.S. bioeconomy from failure to sufficiently provide the attributes, resources, and environment that are necessary to allow it to flourish—failure to promote the bioeconomy. The second section addresses failure to protect the bioeconomy from intentional acts that could harm it, such as theft of intellectual property (IP) or datasets, conferring a competitive advantage on the recipients of that illicitly gained information. It also addresses failure to protect from harms mediated by the bioeconomy that relate to its subversion or misuse, including such traditional biosecurity risks as the development of biological weapons agents, as well as means by which attackers could hijack entities within the bioeconomy to pose risks to people, agriculture, and the environment or to threaten U.S. national and economic security more generally. It is important to note that the committee did not prioritize or rank the risks identified in this chapter, and that while the committee strove to be as comprehensive as possible in outlining risks to the bioeconomy, this chapter should not be taken as providing a comprehensive list. Moreover, as discussed in the introduction to Part III, although this chapter addresses risks associated with the failure of certain aspects of the bioeconomy, that could leave the United States vulnerable to coercion or monopoly pricing from foreign suppliers, it does not quantify or analyze these risks in depth. In the latter case, U.S. consumers might still benefit, but U.S. producers would lose “first mover advantages” and leadership in the relevant technologies. The chapter ends with conclusions.

FAILURE TO PROMOTE THE BIOECONOMY

Risks related to failure to promote the bioeconomy include insufficient U.S. government research and development (R&D) investment, asymmetric research constraints, an inadequate workforce, an ineffective

or inefficient IP environment, and an ineffective or inefficient regulatory environment.

Insufficient U.S. Government Research and Development Investment

As explored in Chapters 3, 4, and 5, a history of strong and sustained U.S. government investment in the life sciences, in computing and information sciences, and in engineering has powered the development of today’s world-leading bioeconomy. To retain this world leadership position, the United States will need to sustain its investment in basic research and the development of supporting and enabling technologies. The committee identified the potential risks described below should U.S. investment in R&D be insufficient.

Loss of Scientific Leadership

Insufficient support for fundamental research, whether from the U.S. government or from major nongovernmental funders, will erode the United States’ ability to achieve breakthrough scientific results, as well as the type of incremental learning that can also have direct economic application. In the longer run, insufficient research support will erode the United States’ ability to develop and recruit the world’s best research talent, including domestic talent, particularly in competition with other countries that are investing heavily in their own bioeconomies (as discussed in Chapter 4). Specifically, loss of U.S. scientific leadership could have the following consequences:

• Significant developments that drive innovation and economic returns could increasingly happen outside the United States.
• Students and researchers who seek the opportunity to work with the world’s best researchers could leave or be less likely to come to the United States, depriving the nation of their expertise.
• Start-ups and other corporations that are formed to build on the scientific advances realized through R&D and that are staffed by the researchers, students, and technologists who have worked with influential academic research groups could be less likely to thrive within the United States. Although research results that are published in the open literature are available anywhere in the world, the existence of biotech innovation clusters, such as those in the San Francisco Bay and Boston areas, shows that there is value to founding a biotech company close to major research institutions and in the vicinity of other biotech firms (Audretsch and Feldman, 1996; Bailey and Montalbano, 2017; Feldman and

• Massard, 2002; NASEM, 2017b). Proximity to scientific leaders matters, as does the rapid transfer of tacit knowledge and learning from peers and competitors.

• U.S. researchers and institutions could be less able to participate in the establishment of global norms, practices, and ethical standards that reflect U.S. values.

Insufficient Development of Enabling Tools, Technologies, and Standards

Investments in basic research have historically led to new applications, even more so when the research has led to the development of a tool or technology that spurred greater innovation in related applications, as has been the case for enabling technologies such as DNA sequencing, DNA synthesis, genome-editing tools, high-performance computing, and data-sharing platforms. Continued funding and support for research that could extend and improve these tools or result in a new enabling technology is paramount to maintaining scientific leadership; however, identifying what research to fund is a perennial challenge. Within the synthetic biology community, the Engineering Biology Research Consortium (EBRC), a nonprofit public–private partnership dedicated to advancing the engineering of biology, has developed a technology roadmap to identify priority areas of precompetitive research over the next two decades (EBRC, 2019). This roadmap, and others like it, can be used by U.S. government programs to focus their investments on precompetitive research topics that will accelerate large segments of the field as a whole. The EBRC roadmap focuses on four technical areas—engineering DNA, biomolecular engineering, host engineering, and data science—highlighting the potential of technical developments in these domains to enable rapid advances across a number of application sectors, including food and agriculture, health and medicine, energy, industrial biotechnology, and environmental biotechnology.

It is worth noting that data-sharing capabilities have greatly accelerated various scientific discoveries and their downstream applications, as is discussed in Chapter 5. Insufficient support for these efforts has the potential to constrain access to data on which researchers within the U.S. bioeconomy rely and could hamper future efforts to share and combine large datasets more efficiently (Toga and Dinov, 2015).

In addition to supporting fundamental scientific research, U.S. government investments and institutions, such as the National Institute of Standards and Technology, support the development of measurement techniques and standards that may not be profitable for any individual private firm to develop but that benefit the U.S. bioeconomy as a whole by making many U.S. firms more productive. For example, the development

and adoption of a set of standard biological components with reproducible characteristics has the potential to enable interoperability, longer and more complex supply chains, and the generation of more complex products (Galdzicki et al., 2011). The number of registries and databases aiming to catalog and make available standard components is growing.² Insufficient attention to and investment in these underlying technologies, particularly in the face of competition from other nations whose governments are funding such investment, will make the U.S. bioeconomy less competitive.

Asymmetric Research Constraints

Constraints placed on U.S. bioeconomy research laboratories but not on academic competitors overseas can create a competitive disadvantage, whether by limiting or preventing U.S. researchers from conducting certain types of research, limiting access to particular materials or samples, or providing incentives for productive researchers to leave the United States for countries with less stringent regulatory environments.

For example, human embryonic stem cells (hESCs) are currently being used in a number of clinical studies, including those focused on macular degeneration of the retina, diabetes, heart repair, and the induction of T cell–mediated immunity. In the United Kingdom, the Human Fertilization and Embryology Act of 1990 and the Human Reproductive Cloning Act of 2001 permit the destruction of embryos to obtain hESCs for research and treatment of serious diseases (Dhar and Ho, 2009). As a result, the United Kingdom now has a global leadership position in the development of clinical-grade lines suitable for regenerative therapies. In contrast, the U.S. regulatory landscape has been much more restrictive than that of not only the United Kingdom but also, for example, Japan and Singapore (Dhar and Ho, 2009). Following an outright ban in 1995 on the destruction of human embryos for research, the restrictions were relaxed in 2009 to allow the generation of new human embryonic cell lines, with a number of ethical provisions involving donor consent.³ More than 100 lines in the National Institutes of Health (NIH) hESC registry that carry specific mutations linked to monogenic diseases, such as cystic fibrosis and Huntington’s disease, were generated but not widely utilized for research because of ethical issues, the limited number of diseases involved, and the regulatory landscape (Ilic and Ogilvie, 2017). An analysis of the research literature shows that the U.S.-based share of worldwide research into

___________________

² iGEM Registry of Standard Biological Parts (http://parts.igem.org/Main_Page); the Synthetic Biology Open Language (http://sbolstandard.org and https://doi.org/10.1016/j.synbio.2018.04.002); see also Feuvre and Scrutton (2018).

³ See https://doi.org/10.1038/ncb0710-627.

hESCs is decreasing, while the work of Chinese groups—which have not faced the same constraints—is increasingly being published (Guhr et al., 2018). The growing performance of Chinese groups in hESC research may be an immediate consequence of extensive funding programs and strong political support (Guhr et al., 2018).

Induced pluripotent stem cells (iPSCs) can be derived from adult somatic cells, as described in Chapter 1, and may eventually obviate the need for hESCs for drug discovery, as disease models, and for cellular therapies to cure disease. Because iPSCs are derived from adult cells, however, these lines have acquired genetic mutations and epigenetic modifications over the lifetime of the cell donor that may impact their clinical utility. Thus, hESCs remain the “gold standard” for what may be possible for cellular therapies using iPSCs in the future (Ilic and Ogilvie, 2017), and the majority of current clinical trials are based on hESC-derived cell products (Guhr et al., 2018). In addition to companies conducting trials in the United States, companies in Brazil, China, France, Korea, and the United Kingdom are at the forefront of clinical translation in this arena.

Additional examples of regulatory research constraints include regulations limiting the use and types of animals for research purposes and restrictions related to the use of particular pathogens.

Inadequate Workforce

Growth of the U.S. bioeconomy may be hindered if the quantity or quality of workers with the appropriate skills is insufficient to meet demand. Not only is a skilled workforce necessary to supply U.S. bioeconomy firms with the best possible talent, but a high-quality technical workforce can provide an incentive for foreign bioeconomy firms to establish research and production facilities in the United States.

The ability of the U.S. K–12 education system to engage and prepare students to study science, technology, engineering, and mathematics (STEM) subjects at the university and postgraduate levels has long been of concern. Many studies have offered recommendations for improvement, including improving outreach to minority-serving institutions, devising new mechanisms for undergraduate students to participate in research, and taking part in such programs as the International Genetically Engineered Machine (iGEM) competition (see Box 7-1).⁴

___________________

⁴ Among the many reports of the National Academies calling attention to the need to strengthen the U.S. STEM workforce are Rising Above the Gathering Storm (NAS et al., 2007); Rising Above the Gathering Storm, Revisited: Rapidly Approaching Category 5 (NAS et al., 2010); Undergraduate Research Experiences for STEM Students: Successes, Challenges, and Opportunities (NASEM, 2017d); Graduate STEM Education for the 21st Century (NASEM, 2018b); Indicators for Monitoring Undergraduate STEM Education (NASEM, 2018c); and Minority-Serving Institutions: America’s Underutilized Resource for Strengthening the STEM Workforce (NASEM, 2019).

U.S. colleges and universities can improve the number and quality of their technical graduates, researchers, and educators by continuing to attract high-quality science and engineering students and scholars from overseas. Foreign students constitute a significant fraction of the enrollments at U.S. colleges and universities, particularly in STEM disciplines, and foreign-born employees form a substantial component of the U.S. STEM workforce.⁵ Both domestic and international factors may complicate the ability of the United States to continue to attract scientists and engineers to this country.

___________________

⁵ In the field of biological, agricultural, and environmental sciences, foreign-born scientists and engineers constituted 15.4 percent of the workforce with a bachelor’s degree; 27.3 percent of those with a master’s degree; and 46.9 percent of those with a Ph.D. in 2015. Note that “foreign-born” is a broader category than individuals who initially arrived in the United States on a (temporary) student or scholar visa; it includes foreign nationals who have immigrated to the United States in any capacity and have attained permanent residency or citizenship (NSB and NSF, 2018).

Internationally, opportunities for students to remain in their home countries are growing as foreign bioeconomies expand. The world’s best science and engineering students and scholars have an increasing number of options for where to study and do research other than coming to the United States. As Federal Bureau of Investigation (FBI) Assistant Director Edward William Priestap testified before a Senate Judiciary Committee subcommittee in June 2018, “Any research institution hoping to be—and to remain—among the best in the world must attract and retain the best people in the world, wherever they are from” (DOJ, 2018b, p. 5). Assistant Director Priestap also called attention to the risk that “some foreign actors, particularly foreign state adversaries, seek to illicitly or illegitimately acquire U.S. academic research and information to advance their scientific, economic, and military development goals.” He continued by observing that, “through their exploitative efforts, they reduce U.S. competitiveness and deprive victimized parties of revenue and credit for their

work” (DOJ, 2018b, p. 2). A more detailed discussion of this concern can be found later in this chapter.

Domestically, the United States is increasingly restricting the entry of foreign scholars and students into the country by applying visa controls, which regulate temporary visits and permanent immigration by foreign nationals. The degree of scrutiny applied to visitors depends, among other things, on whether their country of origin poses national security concerns, including the intent to seek illicit access to U.S. technology. Visa controls thus enable the U.S. government to deny access to individuals thought to be supporting such hostile state efforts. However, restrictive visa policies applied to classes of foreign nationals also may have the effect of discouraging the participation of foreign students and scholars in the U.S. bioeconomy research community and workforce more generally, whether as a result of the restrictions themselves or the creation of a perception that the United States is hostile to such engagement. On June 3, 2019, for example, the Chinese government warned students that visas to the United States were increasingly being delayed, denied, and restricted, and the next day warned potential tourists that U.S. law enforcement agencies were “harassing” travelers from China (Zheng, 2019a,b). The same month, Massachusetts Institute of Technology (MIT) President Rafael Reif warned against allowing concerns over academic espionage, well-founded as they might be, to create a “toxic atmosphere of unfounded fear and suspicion” that would send the message that the United States “no longer seek[s] to be a magnet for the world’s most driven and creative individuals” (Reif, 2019).

Independent of recent policy changes regarding security screens for foreign students and scholars, U.S. immigration law mandates that applications for student or scholar visas be rejected unless applicants can prove that they have ties to their native country sufficient to compel their return after their U.S. stay. In other words, as stated in a white paper by the Center for Strategic and International Studies, despite the potential contributions that foreign students and scholars can make should they remain in the United States, “the only way they can enter the United States in the first place is by proving their intent to make those contributions somewhere else” (CSIS, 2005, p. 14). It may therefore be difficult to rely on foreign technical expertise to fill gaps in the U.S. bioeconomy workforce.

Ineffective or Inefficient Intellectual Property Environment

Uncertainty over what is considered patentable could have a destabilizing effect on the U.S. bioeconomy by negatively affecting both those pursuing patent protection and those wishing to bring innovations in biotechnology to practice. Since recent Supreme Court decisions have narrowed what is considered patent eligible (discussed below), companies

have experienced more difficulty in obtaining and defending patents on biological innovations. Because patent eligibility is an important consideration for venture capitalists and private equity investors, the greater uncertainty over patent eligibility makes it less likely that firms will invest in biotechnology companies (Taylor, Forthcoming).

Under U.S. patent law, there are two criteria for patent subject-matter eligibility—one statutory, the other judicial. For a claimed invention to qualify as patentable subject matter, it must fall into one of the four statutory categories, defined under 35 U.S.C. § 101 as “any new and useful process, machine, manufacture, or composition of matter.” The claimed invention also must not fall into one of the judicial exceptions created through a series of court decisions, namely, abstract ideas, laws of nature, and natural phenomena (including products of nature) (see Manual of Patent Examining Procedure § 2106.04).

In recent years, a number of U.S. Supreme Court decisions have expounded upon the judicial exceptions to patent subject-matter eligibility, including Mayo Collaborative Services v. Prometheus Labs (566 U.S. 88; Mayo) in 2012, Association for Molecular Pathology v. Myriad Genetics (569 U.S. 576; Myriad) in 2013, and Alice Corporation v. CLS Bank International (573 U.S. 208; Alice) in 2014. In Myriad, the Court held that

In Mayo, the Court held that “Prometheus’ patents set forth laws of nature—namely, relationships between concentrations of certain metabolites in the blood and the likelihood that a dosage of a thiopurine drug will prove ineffective or cause harm,” and therefore were not patent-eligible (566 U.S. at p. 8). In Alice, the Court affirmed the Mayo decision by providing a two-step test of patent eligibility: (1) “determine whether the claims at issue are directed to [a] patent-ineligible concept” (573 U.S. at p. 2) and (2) if the answer is yes, then “search for an inventive concept—i.e., an element or combination of elements that is sufficient to ensure that the patent in practice amounts to significantly more than a patent upon the ineligible concept itself” (573 U.S. at p. 7).

In response to these and other decisions handed down by the U.S. Supreme Court, the U.S. Patent and Trademark Office (USPTO) has continually updated its criteria for evaluating patent subject-matter eligibility (Bahr, 2016, 2018a,b; USPTO, 2014, 2015). In 2017, it issued a formal report on patent-eligible subject matter summarizing the case law, international

approaches to defining patent-eligible subject matter, and the public’s view on patent subject-matter eligibility (USPTO, 2017). The most recent guidance on patent subject-matter eligibility was issued in 2019 (USPTO, 2019a,b).

Changes to USPTO examination practice in response to these Supreme Court decisions have had a substantial impact on patenting in biotechnology. Since the Myriad decision, patent examiners have been narrowing pending patent claims involving nucleotide sequences not only for applications involving human genomic DNA but also for those covering agricultural products (Jefferson et al., 2015). In response to Myriad-based rejections, patent applicants are not “drafting around” the legal principles in Myriad; instead, about half (47.6 percent) are abandoning their claims, and about half (47.9 percent) are amending their claims to overcome the rejections (Aboy et al., 2017). Notably, the Myriad decision is having a broader impact on biotechnology patent applications beyond those involving isolated genomic DNA. Over a 5-year period after the Myriad decision was issued, 6,785 patent applications in Technology Center 1600 (the technology center that provides examination for patent applications in biotechnology and organic chemistry) received a Myriad-based rejection, 85 percent of which covered products other than naturally occurring DNA (Aboy et al., 2018).

The Mayo decision also has had a substantial impact on patenting in biotechnology. An analysis of patent applications filed in Art Unit 1634 (an art unit responsible for a substantial number of biotechnology inventions) found an increase from 10.5 percent (pre-Mayo) to 55.5 percent (post-Mayo) in applications that were rejected for not satisfying the patentability conditions in 35 U.S.C. § 101 (Aboy et al., 2019). Even higher rejection rates were observed for patent applications focusing on personalized medicine—an increase from 15.9 percent (pre-Mayo) to 86.4 percent (post-Mayo) in 35 U.S.C. § 101 rejections (Chao and Mapes, 2016). Among the broader collection of patent applications filed in Technology Center 1600, fully 4,650 (49.3 percent) of applications receiving a Mayo-based rejection in the 6 years after Mayo was decided were abandoned (Aboy et al., 2019). In addition, Mayo has substantially increased the time and costs for prosecuting patent applications in biotechnology. Among the subset of patent applications in Technology Center 1600 that were able to overcome a Mayo-based rejection, 45.8 percent had to file one or more Requests for Continued Examination, and 30.3 percent had to file two or more such requests (Aboy et al., 2019).

These decisions also impact granted U.S. patents that are challenged in court. As an example, in Ariosa Diagnostics v. Sequenom (788 F.3d 1371 [Fed. Cir. 2015]; Ariosa), the Federal Circuit Court affirmed the District Court’s finding that the claims of the patent in question are not directed to patent-eligible subject matter and are therefore invalid under 35 U.S.C.

§ 101. The patent at issue in Ariosa concerned detecting cell-free fetal DNA in maternal plasma to identify fetal characteristics and abnormalities, an invention that replaces invasive prenatal techniques. Using the two-part § 101 test, the Court found (1) that the claims “are directed to a patent-ineligible concept” because the “method begins and ends with a natural phenomenon” (i.e., cell-free fetal DNA), and (2) the claimed method does not “‘transform’ the claimed naturally occurring phenomenon into a patent-eligible application” of the phenomenon. The Court did not disagree that “detecting cell-free fetal DNA in maternal plasma or serum that before was discarded as waste material is a positive and valuable contribution to science,” but found that “even such valuable contributions can fall short of statutory patentable subject matter.”

These findings reveal an unusually high degree of legal uncertainty both in prosecuting patent applications and in upholding the validity of granted patents in biotechnology. And while it is possible to overcome rejections under 35 U.S.C. § 101, doing so requires time and money. Thus, start-up companies with smaller budgets and limited access to patent expertise are more at risk relative to larger, well-established companies.

Although the empirical data collected to date do not provide conclusive evidence that § 101 should be amended, draft legislation to reform § 101 and other sections of the Patent Act has been proposed.⁶ The proposed legislation seeks to base patent eligibility on the usefulness of the invention, which is defined to be “any invention or discovery that provides specific and practical utility in any field of technology through human intervention.” In essence, the proposed legislation would abrogate the Supreme Court’s two-part § 101 test; eliminate judicial exceptions to patent eligibility; and draw strict lines between the inquiries of §§ 101, 102, 103, and 112. A series of public hearings before the U.S. Senate Subcommittee on Intellectual Property featured testimony from a former chief judge of the U.S. Federal Circuit Court, inventors, industry executives, law professors, former directors of USPTO, and such groups as the American Civil Liberties Union. Over the course of these hearings, the lack of consensus on whether the proposed legislation or other reform of U.S. patent law would help or harm innovation in the life sciences and biotechnology became clear. Notably, a letter signed by more than 80 well-established and respected U.S. scientists, including a number of Nobel laureates and recipients of the U.S. National Medal of Science, urged Congress “to perform a thorough study of the nation’s requirements for patent eligibility

___________________

⁶ Senators Tillis, Coons, Collins, Johnson, and Stivers, Draft Bill to Reform Section 101 of the Patent Act, released May 22, 2019, available at https://www.tillis.senate.gov/services/files/E8ED2188-DC15-4876-8F51-A03CF4A63E26.

and of the draft proposal’s potential consequences for our country’s science and industry, before enacting any relevant legislation.”⁷

The constitutional purpose for granting patents is to promote the progress of science and the useful arts (U.S. Const. Art. I, Sec. 8, Cl. 8). Ultimately, the patent system needs to strike a balance in granting exclusive rights that will encourage innovation while not obstructing access to the fundamental tools of science and biotechnology that should be available to all.

Ineffective or Inefficient Regulatory Environment

Excessive or poorly designed regulations could impede innovation by constraining the choices available to innovators or imposing on them requirements that would tend to increase cost or uncertainty. On the other hand, to the extent that regulations are perceived as protecting public health, public safety, and the environment, they can strengthen public trust in a new technology, leading to wider public acceptance and serving as an innovation driver. Where regulations set a high standard of performance that a regulated product must meet, they can also drive the innovation necessary to meet that standard. An example is fuel economy standards for motor vehicles, which have stimulated innovation in improving fuel efficiency.⁸

However, uncertainty in the regulatory environment, more than the regulations themselves, can serve as a drag on innovation. If innovators know what is expected, they can consider regulatory requirements along with other requirements a new product must be designed to satisfy, such as customers’ cost and performance targets. But if the regulatory environment is uncertain, an innovator may not know which approach to pursue, and may be reluctant to invest too much R&D funding in areas that might be precluded by later regulatory changes. Uncertainty in the regulatory environment can also discourage innovation by encouraging developers to imitate products that have already charted a path through the regulatory system instead of pursuing innovative products that may have unknown paths with long regulatory delays. The 2016 National Strategy for Modernizing the Regulatory System for Biotechnology Products⁹—intended to reduce regulatory uncertainty by clarifying the roles and responsibilities of current regulatory bodies—and

___________________

⁷ See https://www.patenteligibility.com.

⁸ See https://www.transportation.gov/mission/sustainability/corporate-average-fueleconomy-cafe-standards.

⁹ See https://obamawhitehouse.archives.gov/sites/default/files/microsites/ostp/biotech_national_strategy_final.pdf.

the 2019 Executive Order on Modernizing the Regulatory Framework for Agricultural Biotechnology Products¹⁰ represent recent attempts to further streamline the regulatory process. To inform efforts to reduce uncertainty, such studies as the National Academies report Preparing for Future Products of Biotechnology (NASEM, 2017a) can give the regulatory system advance warning of innovations that may not fit comfortably within existing regulatory paradigms. It will be important for the regulatory system to continue to track the progress of innovation in the sectors it regulates, and to ensure that it has developed risk assessment procedures and acquired the resources necessary to be able to develop and implement any necessary regulations without unduly constraining the field.

Lack of Public Trust or Conflict with Public Values

A risk to the U.S. bioeconomy of a very different nature derives from societal factors. In recent decades, societal acceptance, expressed either directly by civil society or through the marketplace, has become a potent determinant of which technologies enter practice and which products survive in the market. Full development of the U.S. bioeconomy will be impaired if its products and services fail to win public trust and acceptance or face opposition. Lack of acceptance or opposition can arise from a wide range of concerns, some of which are discussed in this chapter, while others have been articulated in other venues. These concerns include

• the safety, environmental, or land-use implications of the use of genetic engineering in agriculture or of the production of crops for biofuels;
• the consequences of the release or potential release of genetically engineered organisms into the environment;
• lack of confidence in government regulatory bodies;
• the price of biotechnology-derived medical therapies;
• the distribution of economic benefits between producers and consumers, or among producers of different sizes;
• the distribution of economic benefits between those who generate economic value from genetic information and those who had sovereignty over the specimens from which that genetic information was originally obtained;
• the ethics and propriety of modifying human DNA;
• the ethics and propriety of engineering other living organisms;

___________________

¹⁰ See https://www.whitehouse.gov/presidential-actions/executive-order-modernizingregulatory-framework-agricultural-biotechnology-products.

• the application of biotechnology to human reproduction, including the modification of DNA of future generations;
• propagation of misinformation on the Internet that can put public health at risk (see Box 7-2);
• violations of personal privacy due to unauthorized release of one’s own genetic information;
• violations of personal privacy due to release of one’s relative’s genetic information;
• the degree to which risks that might arise from any given biotechnological activity are borne by the beneficiaries of that activity; and
• the potential use of biotechnology by those deliberately seeking to inflict harm.

Some of these concerns can be addressed by science-based assessments to help determine and convey risks of proposed approaches relative to a range of other risks faced by society, including those of not acting. Such assessments can be used to inform regulatory approaches for risk mitigation. However, “a purely technical assessment of risk could result in an analysis that accurately answered the wrong questions and was of little use to decision-makers,” to quote one National Academies report summarizing another (NASEM, 2016b; summarizing NRC, 1996). Moreover, quantitative assessments may not even address underlying ethical or social concerns or value conflicts that may be crucial to public acceptance and could potentially be addressed through various engagement

strategies (NASEM, 2016a). There obviously are not right or wrong answers to such questions, but rather a spectrum of viewpoints based on the experiences and values of individuals.

The committee recognizes that public acceptance will be important to the development of the bioeconomy and the realization of its potential benefits. However, public acceptance cannot be addressed at the level of the bioeconomy as a whole. Each product, service, or technological innovation developed by the bioeconomy, like products, services, and innovations arising through other types of activity, will be judged by the public on its own merits, through mechanisms and public engagement approaches that will depend on the particular application involved.

FAILURE TO PROTECT THE BIOECONOMY OR TO PROTECT FROM HARMS MEDIATED BY THE BIOECONOMY

In addition to harms done to the U.S. bioeconomy by the nation’s failure to actively promote and support it, the bioeconomy is vulnerable to harm as a result of unfair or illegitimate actions of others, such as the theft of IP, that can harm its competitiveness. Moreover, subversion or misuse of entities within the bioeconomy can cause harm through the accidental or deliberate production and release into the environment of dangerous biological organisms or the corruption of ostensibly beneficial services. As the goods and services offered by the bioeconomy become more widely integrated into the society and the economy at large, adversaries may cause harm through interruption or corruption of bioeconomy operations. Dangerous biological outcomes may be generated through such means as the covert adulteration of biological outputs. And given that the bioeconomy produces goods and services, such as therapeutics and vaccines, that are critical to national security, public health, and public safety, interruption or denial of those goods and services can also lead to societal harm. A healthy bioeconomy must be protected from risks to itself and from the harms that it may pose to the greater society through its subversion or misuse. These risks and harms are discussed in detail in this section.

Constrained Access to International Data

One of the critical inputs for the bioeconomy is data, particularly given the increasing importance of information science, data analysis, and machine learning as a component of the life sciences research process (see Chapter 5). The ability to generate, validate, and use data can be an important source of competitive advantage for biotechnology firms. If foreign datasets are denied to the U.S. bioeconomy as a whole while

foreign entities are able to access U.S. datasets, this lack of reciprocity puts the U.S. bioeconomy at a competitive disadvantage. The same holds true if U.S. firms are forced to release critical bioeconomy datasets to foreign firms as the price of doing business abroad or following a firm’s acquisition by foreign entities.

Asymmetric Access to National Sources of Genetic Information

The U.S. government has enabled and supported the creation of rich information databases relevant to the bioeconomy, such as those containing genomic and other “omics” data, remote-sensing data, research publications and their associated raw data, patent data, and census data. To maximize utilization of the results of publicly funded R&D, the U.S. government’s “open science” initiatives have sought to ensure the public availability (Van Noorden, 2013)—subject to personal privacy protection—of data maintained by the government or developed through government-funded research. However, this approach is not necessarily emulated by other nations that may have amassed similar databases but are not making them available internationally. In addition, the ability of firms, such as BGI in China, to provide very low-cost DNA sequencing allows them to compete for DNA sequencing contracts from U.S. health care providers or to sequence DNA from clinical samples that are sent to associated Chinese firms for analysis. Should these firms retain (or develop and retain) DNA sequence information from U.S. samples, they would amass a dataset of genetic information from the United States whereas U.S. firms would have no way of accessing a reciprocal dataset given the strict regulations on exporting Chinese genetic data or samples (elaborated on below).

Concerns about asymmetric data access are best articulated in the biomedical arena. An increasing number of efforts are under way in research institutions of all types to sequence the genomes of large portions of the human population in order to gain further insights into disease. Examples include the Cancer Genome Atlas Program of the National Cancer Institute (Cancer Genome Atlas Research Network, 2013); the All of Us Research Program and other national efforts (reviewed by Stark and colleagues [2019]); and the work of private companies such as 23andMe, ColoGuard, and Ancestry.com. The private sector is amassing some of the largest datasets. In 2017, the 23andMe consumer database was used to identify 15 genetic loci associated with depression by obtaining the medical records of 400,000 of the firm’s consumers (Hyde et al., 2016). Such achievements exemplify the promise and value of having large, aggregated genomic datasets and the analytic capacity to turn these data into a future product.

However, several countries have enacted policies to prohibit the export of genetic information about their citizens. In 2007, for example, Russia banned the export of all human biological materials, including hair, tissue, and blood, purportedly because the government feared that Western states were developing genetic biological weapons (Vlassov, 2007). Since 2017, Russia has restricted, but not entirely banned, human tissue exports (Bavasi et al., 2017). China does not permit foreign researchers to conduct research involving human genetic resources (genetic materials in human samples or genetic information) unless they are collaborating with a Chinese partner, and the research must be approved in advance by the Human Genetic Resources Administrative Office (Bavasi et al., 2017, p. 2). In 2016, the European Union enacted the General Data Protection Regulation, which expanded health-related data to include genomic and biometric data as “sensitive personal data.” This new regulation requires more detailed informed consent to use an individual’s data for a secondary purpose unless it has been anonymized. Regarding transnational sharing, the regulation requires that the recipient of the data uphold the same standard of data protection outlined by the regulation (Shabani and Borry, 2018). Brazil has adopted a similar framework that requires additional security measures for sensitive personal data and also has extraterritorial reach (Monteiro, 2018). The United States has not enacted comparable policies at the national level and is therefore directed by a series of guidelines and rules (Majumder, 2018). Given the complexities around data sharing associated with differing regulations, it is unsurprising that transnational data-sharing initiatives are being actively developed to ensure continued access (Fiume et al., 2019).

The impact of these regulations on research is yet to be determined. From a public health perspective, banning the export of genetic information from a country would prevent international scientists from conducting research on genetic diseases that were specific to residents of that country, to the detriment of that country’s citizens. From an economic perspective, however, the situation is more complicated. Differences in data protection requirements and ability to share data across the international stage engender concerns about an uneven playing field. If foreign researchers and companies have access to their own countries’ biological datasets as well as to corresponding U.S. bioeconomy data, the larger overall amount of data will give them a distinct advantage in identifying genetic disease mechanisms over U.S. researchers and companies, which would have access only to the latter. While the ethnic and racial diversity of the U.S. population may mean that the U.S. data are more valuable—per patient—for the purpose of global pharmaceutical development than data from countries with more homogeneous populations (Gryphon Scientific and Rhodium Group, 2019), this asymmetry still contributes to an

uneven playing field. In addition, asymmetries in access to data may be compounded if other countries have more permissive regulations around how genomic and clinical datasets can be used.

These asymmetries could allow those foreign companies with more extensive datasets to develop therapies before their U.S. counterparts, enabling them to patent and market those therapies first. Again, strictly from a public health perspective, such outcomes—if the therapies could obtain U.S. Food and Drug Administration (FDA) approval—could be seen as advantageous to the United States, whose citizens would benefit from earlier access to therapies than they would have if they had to wait for U.S. firms (with their lesser data sources). Economic and national security problems could arise in the long run, however, if U.S. manufacturers were consistently scooped in their ability to develop their own products, consequently losing profits and market share. If U.S. firms suffered losses to the point where they were unable to stay in business, the U.S. health care system would find itself dependent on foreign pharmaceutical manufacturers for these products, possibly leaving the nation vulnerable to monopoly pricing or even coercion. The U.S. government also possesses databases that are not open to the public in their entirety but can be accessed, often in redacted form, by researchers with appropriate authorization. Such databases include medical records of those individuals for whom the government provides or finances medical care; they also include census information that is available to the public under the condition that information specific to identifiable people or entities be excluded.

The value of these databases to the U.S. and other national bioeconomies, the vulnerability of these databases to access or exploitation, and the effect a country’s policy on data openness can have on the relative standing of its own bioeconomy all warrant further scrutiny.

Constraints on Genomic Data as a “Genetic Resource” Under the Nagoya Protocol to the Convention on Biodiversity

The United Nations Convention on Biological Diversity (CBD) has initiated discussions on the relevance of “digital sequence information” to the Convention’s goals.¹¹ This move reflects the changing nature of mechanisms for distributing knowledge or information about a biological entity, which traditionally has relied on the exchange of physical specimens but now may be accomplished by generating and distributing various digital representations of those specimens. The most widely discussed digital representation is an organism’s genetic sequence. However, outcome 2 of

___________________

¹¹ See https://www.cbd.int/abs/dsi-gr.shtml.

the Ad Hoc Technical Expert Group on Digital Sequence Information on Genetic Resources¹² illuminated the breadth of what may be considered under this “placeholder” term, which included, among other things, the following (list excerpted from Annex to CBD/SBSTTA/22/2):

1. The nucleic acid sequence reads and the associated data;
2. Information on the sequence assembly, its annotation and genetic mapping. This information may describe whole genomes, individual genes or fragments thereof, barcodes, organelle genomes or single nucleotide polymorphisms;
3. Information on gene expression;
4. Data on macromolecules and cellular metabolites;
5. Information on ecological relationships, and abiotic factors of the environment;
6. Function, such as behavioural data;
7. Structure, including morphological data and phenotype;
8. Information related to taxonomy;
9. Modalities of use.

Given that the information enumerated above resides in various public and private data repositories, the question of equitable access and fair distribution of the economic value derived from that information is at the heart of the current discussion on digital sequence information (DSI) with respect to access and benefit sharing. Lai and colleagues (2019) provide a brief overview of the implications of this access to the growing field of synthetic biology. They conclude that policies regarding DSI “could have a significant influence on synthetic biology research and development internationally. For example, implementation of active ABS [access and benefit-sharing] policies on genetic information could inhibit global commercialisation of public-funded research or promote ‘get-arounds’ to avoid ABS, both of which are not ideal scenarios.” Hiemstra and colleagues (2019) provide stakeholder input on the implications of regulating digital sequence information for innovation in multiple biological and ecological domains from the Dutch perspective. They examine the domains of plant and animal breeding, biological research, human health, and use of microorganisms and the field of biotechnology. One of their examples is the widespread use of enzymes in the food industry, arising from diverse sequences derived globally. The authors argue that challenges in trying to track the origin or redistribution of such sequences would be impossible to overcome, and that mandating such efforts would adversely

___________________

¹² See https://www.cbd.int/doc/c/704c/70ac/010ad8a5e69380925c38b1a4/sbstta-22-02-en.pdf.

impact biotech start-ups and dampen innovation. They conclude that “ABS arrangements for DSI [digital sequence information] would result in an unforeseeable administrative burden, which consequently leads to large costs, delays in research and slowing down of scientific progress and innovation.” Interestingly, they found that Dutch stakeholders felt that “the value of individual genetic resources or DSI is over-rated or overestimated in international discussions. This may result in unrealistic expectations regarding levels of benefit sharing.” In accord with this observation is the finding of an independent study that the use of lactic acid bacteria, which underlies the production of all cultured milk products worldwide, would be adversely challenged by certain mechanisms of implementation of the CBD (Flach et al., 2019). Two notable issues raised are that many of the currently practiced or envisioned mechanisms involve bilateral agreements, which become burdensome if not conflicting, and that with existing practices for the global distribution of such products that themselves contain microorganisms, such as yogurt, these agreements bring with them biological samples that are often isolated, genetically improved, and reused in new products.

Use of Bioeconomy Datasets to the Detriment of Individual Privacy or National Security

Two risks associated with bioeconomy datasets involve harm to either individual privacy or national security: exploitation of genetic vulnerabilities and genetic targeting of populations.

Exploitation of Genetic Vulnerabilities

Whole human genomic data, such as those collected by such companies as 23andMe and Ancestry.com, are building the broader informational dataset about genes, inheritance, and subpopulations. A recent study addresses cybersecurity risks specific to human genomic data, data most relevant to biotechnological manufacturing, and human clinical health metadata (DiEuliis, 2018). The emerging landscape in these domains is one of a continuum of potential harms that range from violations of individual privacy, to individual physical harms, to national security concerns (depending on which individuals or populations are at risk).

It has already been demonstrated that individuals can be identified from even portions of their DNA (Dankar et al., 2018; Erlich et al., 2018), and they can be further identified through DNA information gathered on siblings or close relatives (Cohen, 2018; Kaiser, 2018). This finding has implications for individual privacy, safety, and security. Individuals could be targeted for discrimination or manipulation based on genetic

knowledge, and individual biological vulnerabilities could be targeted for physical harm.

Personal knowledge that might be revealed through analysis of these datasets pertains not only to aspects of disease but also to human attributes and behavior, as genomic studies are revealing the underpinnings of complex behaviors and potential ways to manipulate them. Described as “sociogenomics” (Comfort, 2018; Robinson et al., 2005), this arena represents another category of data that could be used to further the intent to do harm. Information about an individual’s genotypic predilection for disease or phenotypic behaviors could be used for harm in a social context or to promote discrimination or extortion of an individual. Or there could be known ways to exploit a particular genetic vulnerability to harm to an individual. Electronic health records, health insurance profiles, or other clinical databases in which such data may be housed thus represent important resources that merit protection. In the past few years, comprehensive data thefts have been possible through direct cyberattacks on health information technology infrastructure at large health insurance companies (Ellis, 2018; Ronquillo et al., 2018).

These potential harms become national security concerns when they provide adversaries with a means to elicit personal information about, or even mechanisms to influence, key national decision makers or security personnel, such as members of the military or police forces. It may never be possible to associate a genetic trait with a particular decision; nonetheless, the propensities of national leaders to act in certain ways, which could be influenced by their genetic makeup, could well be of interest to adversarial intelligence agencies. Even if genetic associations with behavior are not well understood at present, they will become better established as more research is conducted and more data are collected and analyzed (Braudt, 2018).

Even if an individual of concern has never provided a genetic sample for the purpose of uploading into a commercial genetic or genealogical database, genomic information is increasingly being derived from medical samples in the pursuit of personalized medicine—the tailoring of medical treatments to a patient’s individual characteristics, including genetic makeup. Rapidly decreasing costs for whole-genome sequencing—currently about $1,000 per genome and falling rapidly—are accelerating this trend.¹³ And once any such genome is available in a database, it will remain relevant to that person’s relatives and descendants forever, albeit decreasingly so as the relationships become more distant.

Similar targeting could be performed using plant or animal genomic data as precision agriculture makes use of individualized genomic

___________________

¹³ See https://www.genome.gov/about-genomics/fact-sheets/DNA-Sequencing-Costs-Data.

techniques equivalent to precision medicine. Advances in these fields are just as important as those in precision medicine, and are also a target for exploitation.

Genetic Targeting of Populations

Discussions of national security risks posed by access to genetic databases increasingly involve questioning whether “genetic weapons” might be feasible.¹⁴ Such weapons would confer the ability to attack a specific individual or group of individuals on the basis of distinctive genetic traits that those targets would share but that would be very rare or nonexistent in anyone else. Any genetic weapon would require (1) characteristic genetic sequences that can be found in the genomes of the intended target person or population; (2) the corresponding absence of those characteristic sequences in anybody else; and (3) a biological mechanism—say, a DNA construct delivered by a virus—that, when activated within the body, would become highly pathogenic if, and only if, those characteristic genetic sequences were present.

With respect to the first of the above criteria, the science of forensic genetics shows that individuals can be uniquely identified by their DNA. The promise of precision medicine in tailoring medical treatments to individuals or groups on the basis of genetic characteristics and the ability of genetic testing services to categorize people into “haplogroups” that share common ancestors in their patrilineal or matrilineal lines make clear that groups of people who share some common genetic characteristics are increasingly being identified. Whether those groupings correlate with criteria an attacker might seek to target (racial, ethnic, social, political, national, or ideological) is less certain. The two remaining criteria face some additional challenges to overcome. For example, even when genetic signatures have been identified that tend to occur more often in certain groups than in others, they may not form precise distinctions, and they therefore may identify a larger group than was intended. Lastly, the construction of a biological mechanism that could identify a genetic signature and trigger a pathogenic process would entail additional technical challenges.

In summary, developing a genetic weapon that would be able to target selected groups of people preferentially poses a number of technical difficulties. On the other hand, information about the human genome is

___________________

¹⁴ Any such weapon based on a biological agent would violate the Biological and Toxin Weapons Convention. Discussion of “genetic weapons” is not meant to imply that they would be legally acceptable or even technically feasible yet.

growing rapidly, and new biotechnologies are continually being developed that lower the barriers to mastering various biological processes. As with the other biosecurity concerns discussed later in this chapter, this area of research will require continual monitoring.

Potential for Violation of Personal Privacy, Utilization by Law Enforcement, and Genetic Discrimination

The popularity and availability of direct-to-consumer (DTC) gene-testing kits have soared in recent years, with hundreds of such DTC services becoming available and an estimated 15 million people taking part as of April 2018 (Erlich et al., 2018; Martin, 2018). Although genetic testing provides a wealth of information, concerns remain about the privacy of genetic information. While some of the more popular services, such as 23andMe, are very explicit about their privacy policies (Martin, 2018), most such services are not. A study of the privacy policies of 30 different DTC genetic testing companies found that most “do not consistently meet international transparency guidelines related to confidentiality, privacy, and secondary use of data” (Laestadius et al., 2017).

Additionally, there are risks associated with linking genetic data to personal information that is posted to such databases as GEDmatch. While these public third-party services have classically been used to identify distant relatives by matching genetic information procured from DTC companies, this use does not exclude the ability to use this information for other purposes. And one study showed that 60 percent of individuals with European ancestry in the United States can be linked to at least one individual in the GEDmatch database who is considered a close relative. Recently, law enforcement has been utilizing third-party genetic information websites to identify criminals, predominantly in cold cases (Saey, 2018). Using genetic evidence gathered at crime scenes, law enforcement can find relatives with close genetic ties to criminals and subsequently develop a list of suspects. These suspects can then be confirmed by direct genetic testing (Saey, 2018). The most famous of these cases is the recent identification of the Golden State Killer, who was active between 1974 and 1986 but was identified and arrested in 2018 following the use of genetic genealogy (Jouvenal, 2018). In response to this lack of privacy with respect to law enforcement, GEDmatch adopted a new privacy policy in May 2019 requiring its users to opt in to use of their genetic information by law enforcement (Aldhous, 2019).

The U.S. Department of Defense maintains a repository of DNA reference specimens for all active duty and reserve service members, but a court order is required to release them, and only for the purpose of

“investigation or prosecution of a felony, or any sexual offense, for which no other source of DNA information is reasonably available.”¹⁵

In September 2019, the U.S. Department of Justice adopted an interim policy that establishes requirements for the use of this type of genetic analysis by law enforcement. One requirement is that investigative agencies identify themselves as law enforcement to the genetic genealogy services they use, and that they utilize only genetic genealogy databases that have explicitly notified their users that law enforcement may use their services to investigate crimes or identify human remains. The interim policy also sets out how the practice is to be used to generate leads for unsolved crimes (DOJ, 2019; DOJ Office of Public Affairs, 2019).

Another consequence related to the rise of genomic sequencing is the potential for genetic discrimination. As discussed in the earlier section on genetic targeting of populations, genetic discrimination can be based on genetic characteristics within a group of genomes, as well as individual genetic characteristics. The most prominent example of group discrimination and surveillance is the use of genetic sequencing by China to identify Uighurs, a Muslim ethnic group. Members of this ethnic group were identified by the Chinese government under the guise of health-related genetic testing, but with the purpose of placing them in “re-education camps” to be “more subservient to the communist party” (Wee, 2019).

Genetic testing also allows for discrimination against individuals based on their genetic predisposition to traits or diseases. Congress took action to mitigate this problem in 2008 by passing the Genetic Information Nondiscrimination Act (GINA), which prohibits discrimination by employers and health insurers based on genetic information, but it fails to cover many other critical areas in which discrimination is possible, such as life insurance or health care plans from employers with fewer than 15 employees. Some states passed their own policies to close these gaps. California, for example, enacted its comprehensive CalGINA, covering discrimination in many scenarios, including life insurance and disability insurance.¹⁶

One example of individual genetic discrimination is a child, Colman Chadman, of Palo Alto, California, who had genetic markers for cystic fibrosis (CF) without having the disease. Chadman was attending a school where there were two other children with CF, but was dismissed because of the possibility that multiple children with CF in the same school could enable the possibility of transmitting infections (the other two children were siblings and therefore allowed to stay together in school). His family

___________________

¹⁵ 10 U.S.C. § 1565a, “DNA samples maintained for identification of human remains: use for law enforcement purposes.”

¹⁶ See https://www.genome.gov/about-genomics/policy-issues/Genetic-Discrimination.

subsequently sued for genetic discrimination on the grounds that Chadman had only genetic markers for CF, but not the disease (Zhang, 2016). As genetic information becomes more reliable and reveals more information about individuals, it can open opportunities for new avenues of genetic discrimination.

Another risk is the social instability that could occur if GINA were repealed or weakened. As the economic value and predictive power of information in the human genome increase, certain industries will be able to make increasingly powerful economic arguments for having access to and being able to use human genome information in their decision making.

Cyber Risks Associated with the Bioeconomy

With the increasing reliance on large aggregated datasets, the emerging bioeconomy now exists at the intersection of information science and biotechnological science. The digitization of biology—most literally, the conversion of nucleotide codes of DNA to machine-readable formats—is transforming all the life sciences. DNA sequences can now be databased, mined, and used for in silico experimentation or design. To fully extrapolate digital information into meaningful biological systems or the creation of engineered organisms requires more than representation in machine-readable formats. The leap from the nucleotide data sequences recorded in databases to tangible biological predictive form and function is referred to as “abstraction” (Ochs et al., 2016), and will be enabled only through deeper understanding of how genomic sequence underpins function and phenotype, using a complex set of computational tools, algorithms, and bioinformatics programs. Abstraction would enable a future biological engineer to sit at a computer interface and simply type in desired phenotypic features for a biological protein/enzyme, or even an entire microbe or plant cell, and receive those designs as outputs without directly knowing the genetic sequences responsible for those phenotypes. The more complex the organism is, the greater the computing and data storage power that will be required.

A second important advance is automation, which increasingly drives biological manufacturing platforms—machines can now do much of the work that previously could be accomplished only by human physical handling. Furthermore, automated devices that monitor and/or control biological and physiological processes produce reams of data in highly parallelized sets of experiments, running 24 hours per day, which can be shared and stored through cloud computing networks, and as noted above, the operation of such devices requires advanced computational software, algorithms, and bioinformatics. Moreover, increasing amounts

of data are generated during the monitoring and control of bioeconomy-related commercial manufacturing processes, and it is critical to these commercial enterprises that such data be secured and protected as part of a quality management system (Mantle et al., 2019).

Resources in the bioeconomy are valuable, both commercially and because of the risk to life, national security, health, and property if a malicious party should tamper with, access, or otherwise manipulate the data. For the past 20 years, most malicious hacking has been goal-directed, with financial or national interests as the primary motivators. As Table 7-1 demonstrates, bioeconomy companies are major targets for both of these motivators. Many of the most sophisticated cybersecurity attacks will likely originate from or be abetted by foreign intelligence agencies. Such agencies can bring to bear more technical skills and more resources than can ordinary criminal hackers. These skills and resources include what one former National Security Agency official has called “the three Bs: burglary, bribery, and blackmail” (Smith and Marchesini, 2007). Note that these attacks may specifically target corrupt or coerced employees, that is, people who have authorized access to computer systems and who are inside many firewalls.

The bioeconomy’s growing reliance on software, networking, and computer hardware tools yields the same cyber vulnerabilities present in any other sector, which can be viewed as fundamental cybersecurity risks. Cybersecurity here, as in other sectors and domains, is typically concerned with hacking, sabotage, or other compromise of cyber controls that can result in disruption, breached privacy, or theft of IP. These kinds of activities can have adverse impacts on the bioeconomy, and furthermore, on the U.S. economy writ large. A recent report from the White House estimates that malicious cyber activity imposes costs on the U.S. economy (through the theft of IP and personally identifiable information, denial-of-service attacks, data and equipment destruction, and ransom-ware attacks) that ran as high as $109 billion in 2016 (Council of Economic Advisors, 2018).

Understanding of the security vulnerabilities that may derive from cyber intrusions has recently generated discussion of what is referred to as “cyberbiosecurity” (Murch et al., 2018; Peccoud et al., 2018). Cyberbiosecurity has been described as bringing together “disparate communities to identify and address a complex ecosystem of security vulnerabilities at the interface of the life sciences, information systems, biosecurity, and cybersecurity” (Richardson et al., 2019). Bioinformatics datasets, other input tools or data, or industrial process control systems used by a biotech facility could be vulnerable to tampering, which could result in damage to the facility or the subversion or sabotage of its products, and subsequent harm to people, plants, animals, or the environment

TABLE 7-1 Cybersecurity and the Bioeconomy: A Timeline of Selected News and Events

NOTES: Dates reflect when the incidents were first reported. GE = General Electric; SEC = Securities and Exchange Commission; UCLA = University of California, Los Angeles.

SOURCES: This information was provided to the committee as an early draft of a study undertaken by the Carnegie Endowment for International Peace, conducted by Katherine Charlet (a committee member), Natalie Thompson, and Frances Reuland. See https://carnegieendowment.org/programs/technology/biotechnology/timeline (accessed December 1, 2019).

(Peccoud et al., 2018). Similarly, corruption of environmentally or health-related sensors or data could result in the misapplication of health care or environmental remediation. For example, preventing sabotage of biological containment systems that could cause the environmental or occupational release of certain dangerous pathogens is a required component of security plans for those types of facilities (CDC and USDA, 2017a), but these considerations may not have been evaluated for other components of the bioeconomy that pose similar risks. Given that the security plans of containment labs consider cyber intrusions along with insider threats (CDC and USDA, 2017b), they may offer a useful model for information systems security controls for other bioeconomy components.

The growth of cloud computing and cloud storage will pose new challenges. On the one hand, cloud systems are often inherently more secure because they are administered by specialists. On the other hand, users of these cloud systems need to configure their portion—particularly access controls—properly if security is to be maintained. It is not possible to predict which aspect will dominate, especially if organizations attempt to share some portions of their cloud storage.

Although there is no one model for the use of information systems across the bioeconomy, a few important common features can be identified:

• The bioeconomy relies on large databases, often of commercially or personally sensitive information.
• Some components of the bioeconomy rely on open-source software packages, often of uncertain quality, robustness, and degree of maintenance.
• The bioeconomy relies on Internet communications to exchange data (such as publicly available genome data). Proprietary systems are often used to ensure safety and compliance with applicable regulations for commercial products and processes.

None of these features is unique to the bioeconomy, but their particular manifestation in the bioeconomy is notable. For example, while many commercial datasets involve such personal information as addresses and credit card numbers, the datasets in the bioeconomy (and data being exchanged over the Internet) may include full genetic sequences of humans and other organisms. Arguably, the features described above are thus materially different when understood in the context of the bioeconomy—because the genetic information literally defines us as humans and enables manipulation at the level of life’s component parts. The bioeconomy enables an overlap of privacy risk and the risk of physical harm. As noted earlier, understanding the genetic makeup of an individual can

reveal such vulnerabilities as the propensity for certain diseases and that information could in turn be used to harm a person or group of people.

Within today’s bioeconomy, large corporations are aware of traditional cyber concerns and utilize information technology infrastructure to protect against common threats. However, they may be less aware of the possibility of specific unwanted biological outcomes and their sequelae. Smaller companies or biotech start-ups may not view themselves as cyber targets, or if they do, they may not have the resources to address the risks adequately. Small companies and start-ups are generally more vulnerable to cyber intrusions relative to large organizations. Even if they have skilled information technology departments, such organizations typically have neither the budget nor the security focus to fend off attackers, nor do they have much actual experience in this arena (Hiscox, 2018). They may not employ state-of-the art defenses, such as multifactor authentication, and users who have not been properly educated on these matters are more likely to fall for phishing attacks and the like. In addition, most application programmers have little, if any, education in how to write secure code, opening the door to even low-end attackers.

Addressing cyber concerns also will depend on the commercial availability of mitigation measures. If tools tailored specifically to the biotechnology realm are required, awareness is needed among cybersecurity professionals, who at present have little interaction with bio-specific concerns. Thus, not all of the responsibility for addressing cyber concerns lies in the biotechnology industry and life sciences research space; many cyber-focused programs lack awareness of the particular challenges that research in the life sciences or biotechnology industries may face.

Risks Related to Cyber-Physical Systems

In the bioeconomy, some more novel dimensions of risk beyond fundamental cybersecurity must be considered. These include in particular cyber intrusions that result (whether intentionally or unintentionally) in unwanted or dangerous biological outcomes. Some of these security vulnerabilities have been described previously (Peccoud et al., 2018). One way in which some bioeconomy software, together with associated systems, differs from run-of-the-mill enterprise software is that some of it controls physical devices, such as DNA synthesizers or building services equipment in biological containment labs. Cyber-physical systems pose significant security and safety risks since their compromise can have effects on the real world; in this case, those effects could include faulty or even dangerous synthesis of biomaterials or interference with biological containment systems.

The challenge of securing cyber-physical systems is especially grave because the control computers involved are sometimes running obsolete, unsupported operating systems. Briefly, the lifetimes of the controlled devices (hardware) are often much greater than those of the operating systems (software) on which they rely. As long as the physical functioning of the devices is adequate to the task at hand and they meet any certification requirements, they are typically kept in service. This can be true for many of the devices used in the bioeconomy for research purposes because they are often quite expensive or difficult to change; therefore, discarding them when the operating system or software running on them is obsolete is often not an option. In commercial settings, updating of software or devices because of security concerns is frequently hampered by regulation rather than cost (Williams and Woodward, 2015). In a recent survey of international leaders in biotechnology and cybersecurity, more than 90 percent of respondents expressed the belief that insufficient time and resources were being dedicated to cyber risks to biological equipment and facilities (Millett et al., 2019). If, however, there was a shift in industry practice and the requirements for certification were to emphasize security and appropriate security updates for the lifetime of the device, progress could be made. It is likely impossible, or at best difficult, to retrofit this sort of certification requirement to existing devices, but with lead time, sensible requirements, and well-considered guidance, device manufacturers would be able to comply with new security requirements. There would be costs, but if embedded device manufacturers had to plan for security as a long-term attribute of their products, they would engineer them in such a way as to provide cost-effective lifetime security.

Risks Related to Datasets

Another way in which bioeconomy software is distinct is that some of it operates on very large, very sensitive datasets. Some of these datasets may contain individuals’ genomic or medical data, in which case they entail serious personal privacy risks; others may contain proprietary DNA sequences or other data used to make products that will compete in the marketplace. A variety of operations are performed on these databases, increasingly including use of machine learning and other artificial intelligence techniques that can, for example, associate a protein’s amino acid sequence with its three-dimensional structure or identify pathways for and optimize the production of biosynthesized materials, or—particularly in association with other sources of data such as medical records—“identify the relations between genetic characteristics and the response to specific treatments” or identify new drugs “by training a classifier on a dataset where functioning and nonfunctioning drugs have been

identified” (Oliveira, 2019). Use of these data is vital for the bioeconomy, but they require a great deal of protection. The risk component in this arena is the theft of genomic, medical, or other biotechnological data that could be used to advance a competitor’s efforts or even an adversary’s bioeconomy. In such cases, direct harm to privacy or to an individual may not be the outcome; rather, harm may result from subsequent inappropriate use of the data. Such harms could include the ability to outcompete the United States by inappropriately amassing larger, more comprehensive biotechnology datasets, thus putting the United States at potential economic disadvantage or forcing it to acquire needed products outside its own bioeconomy (as described previously in the chapter).

The integrity of datasets is also a serious issue. To protect them, they could be digitally signed, although there might be difficult questions about the proper public key infrastructure for this purpose. A digital signature, at best, attests that some party believes that certain content is authentic; it does not, however, state that the proper party believes that. Digitally signed datasets are self-authenticating; as such, they can be safely redistributed by other parties or via peer-to-peer mechanisms, such as BitTorrent.

Vulnerabilities Due to Reliance on Open-Source Software

A large portion of the bioeconomy runs on open-source software, often derived from university research projects. Indeed, the U.S. Department of Energy’s (DOE’s) Systems Biology Knowledgebase (KBase)¹⁷ provides a centralized repository of open-source software numbering in the hundreds for web-registered users, to which developers can contribute new tools (Arkin et al., 2018). Sharing of “narratives” by researchers speeds the analysis of data by new users, and new tools are generated using a software development kit that helps ensure compatibility in workflows. Researchers conduct their in silico experiments and analyses within this free and valuable community resource after registering for an account (Arkin et al., 2018). Because the site is extensively curated, KBase itself may be insulated from some vulnerabilities associated with open-source software. While there is no a priori problem with open-source software—indeed, it is a valuable resource for the community—the software industry has learned that simply making code open-source does little or nothing to guarantee its quality, robustness, and security.

Failure to update open-source components included in some large product or system often means that security holes will persist long after the hole has been patched in the upstream packages. Given how popular

___________________

¹⁷ See https://kbase.us/what-is-kbase.

some open-source packages are, many systems that use them can experience common failures (NASEM, 2017c). Furthermore, the security of a codebase is intimately tied to its overall quality: a high percentage of system penetrations are due to buggy code.

Supply chain attacks in the software ecosystem are another risk to the bioeconomy. The provenance of open-source software is often unclear, without an audit trail showing who made which changes, when, and why. Furthermore, there may be no systematic approach for tracking or repairing bugs.¹⁸ These procedural lacunae leave open the potential for vulnerabilities to be deliberately introduced into the software: a malicious party could plant malware in a bioeconomy software package under the assumption that it will someday be used by a bioeconomy company. Although proprietary software would not share the risk that anyone would be free to engineer flaws into the software, it can also pose supply chain risks, not least due to the risk of compromised insiders (Black et al., 2016).

Cybersecurity Protections and Defense in the Bioeconomy

The discussion above describes a number of digitization- and cybersecurity-related risks to the bioeconomy. Fortunately, most of the attacks that can be expected are not as sophisticated as those launched or abetted by intelligence agencies, and can be dealt with via standard, off-the-shelf defensive cybersecurity tools—tools routinely used by many companies. For example, one best security practice is to ensure that all network connections are encrypted. This measure is not so much for confidentiality as for the connection authentication that is part of standard encrypted connections. Similarly, since phishing for user credentials is a ubiquitous attack vector, another best practice is to ensure that all logins (especially for email) are protected via multifactor authentication. That said, more sophisticated attackers do exist, and must be planned for; however, even nation-states tend to try simpler attacks first.

___________________

¹⁸ We note that this is not an inherent problem for open-source software. A number of packages, such as the Apache web server and the Firefox web browser, do use state-of-the-art software engineering practices.

Cyberthreat actors, including foreign intelligence agencies, sometimes pursue broad campaigns not just against one company but against entire sectors. Robust information sharing thus helps spread information that enables companies to take quicker mitigating action to counter these campaigns.

In certain critical infrastructure sectors, Information Sharing and Analysis Centers (ISACs) are key entities in facilitating information sharing.¹⁹ These organizations provide a central place for companies to distribute cyberthreat indicators, receive warnings from government agencies, facilitate training, and act as a cybersecurity resource for the sector.

More recently, the U.S. Department of Homeland Security has encouraged the development of Information Sharing and Analysis Organizations (ISAOs). ISAOs are similar to ISACs in that they provide a forum for sharing information on cyberthreats, but because they do not align with specific critical infrastructure sectors, they can be more flexible in their approach and membership. For example, companies can form a regional ISAO even if they come from diverse sectors.

Companies across the bioeconomy would benefit from participating in a cyberthreat information-sharing organization. However, there is no broadly applicable “fit” for bioeconomy companies within the current structure. Because ISACs are tied to specific critical infrastructure sectors, no single ISAC obviously aligns with the bioeconomy, although some, such as the National Health and Research & Education Network ISACs, would overlap with some portion of bioeconomy stakeholders.²⁰ ISACs vet new members to ensure that they will protect sensitive information that is shared by other members, which means ISAC participation may be more difficult for some members of the bioeconomy, such as start-ups or other companies without much corporate history, than for others.

There are also unique information-sharing needs for the bioeconomy that may not be filled by existing structures. For example, if only the health-focused members of the bioeconomy were sharing threat information with one another, it might be difficult to identify and understand a

___________________

¹⁹ Critical infrastructures are those assets, systems, and networks that are considered so essential “that their incapacitation or destruction would have a debilitating effect on security, national economic security, national public health or safety, or any combination thereof” (see the U.S. Department of Homeland Security website at https://www.dhs.gov/cisa/criticalinfrastructure-sectors).

²⁰ Presidential Policy Directive 21, “Critical Infrastructure Security and Resilience” (White House Office of the Press Secretary, 2013), identifies 16 critical sectors: chemicals; commercial facilities; communications; critical manufacturing; dams; defense industrial base; emergency services; energy; financial services; food and agriculture; government facilities; health care and public health; information technology; nuclear reactors, materials, and waste; transportation systems; and water and wastewater systems.

(hypothetical) adversary cross-sector campaign, involving entities outside health care, to steal bioeconomy-related IP or data. Although it would be possible for bioeconomy stakeholders to form an ISAO, start-up costs are entailed in building such a structure.

It is not feasible to demand that every graduate student research project conform to such standards. Indeed, such standards are uncommon even in computer science departments, let alone biology departments. That said, it would be useful if core open-source bioeconomy software—major programs used by a significant number of companies—were brought into a more formal regime, such as a repository. That is, some version would be captured, audited, and placed under formal change control, with a formal testing regimen and changes restricted to authorized personnel. This process need not and should not change the open-source nature of the software, and anyone would remain free to download it and modify it as desired; changes, though, even those contributed to the package by some user or company, would need to go through an auditing and testing process.

Such a repository could be run by an ISAC-like entity or other special-purpose consortium. Note that it is unlikely that access to the “official” source code within the repository could be restricted to ISAC or consortium members; many open-source packages use the GNU Public License, which bars restrictions on redistribution.

of problem. This approach, though in some theoretical sense unsatisfying, has proved quite successful. Encrypted search—picking out the right records from an encrypted database—often takes only a small integer multiple of the time required for a native database query.²¹ Given how fast today’s computers are, this slowdown is quite acceptable.

Other research has been done on privacy-preserving machine learning, which allows identifying information to be removed from databases while leaving them useful for research (Al-Rubaie and Chang, 2018). This technology may also resolve tensions with some countries about export of their citizens’ data, at least where the objections are rooted in privacy principles and not protectionism. Privacy-preserving machine learning does, however, have limitations, not so much in the algorithms themselves as in the database anonymization process: the effort to protect privacy can obscure crucial details necessary for adequate results (Fredrikson et al., 2014). Further research will be required in this area before wide-scale application can be expected. Furthermore, there are many desired operations for which no MPC algorithms exist.

Economic Attack: Theft and Infiltration

Theft or Misappropriation of Trade Secrets

Theft of trade secrets poses a substantial risk to biotechnology companies. Because of the risks posed by disclosing information in patents (see Box 7-3), many biotechnology companies decide to protect their IP assets as trade secrets instead. As illustrated by Genentech, Inc. v. JHL Biotech, Inc. (No. 3:18-cv-06582-WHA [N.D. Cal. 2019]), trade secrets may be stolen by trusted employees to advance the interests of other parties, including companies outside the United States. In that case, four former employees of the U.S.-based biotech firm Genentech, Inc. were indicted for stealing trade secrets to assist JHL Biotech, Inc., a Taiwan-based company, in developing and manufacturing biosimilar versions of Genentech medicines. The complaint alleges that hundreds of files containing confidential information were downloaded from Genentech’s secure document repository system, including the company’s proprietary, FDA-approved analytical methods; formulation know-how; quality acceptance criteria; and manufacturing protocols and procedures for establishing and maintaining safe, sterile manufacturing facilities and equipment.

In addition to theft of confidential documents, proprietary seeds or strains may be stolen and passed on to other companies. In 2018, for

___________________

²¹ For a summary of encrypted search techniques, see http://esl.cs.brown.edu/blog/howto-search-on-encrypted-data-introduction-part-1.

example, a Chinese scientist who worked as a rice breeder for Ventria Bioscience in Junction City, Kansas, stole genetically engineered rice seeds that expressed recombinant human proteins (DOJ, 2018a).

Illicit Transfer of Knowledge and Technology via Academic Misconduct

The U.S. government has recently become concerned about inappropriate actions taken by foreign students and scholars in U.S. research institutions. In congressional testimony, FBI Assistant Director Priestap stated that U.S. academic environments offer “valuable, vulnerable, and viable targets for foreign espionage” that are exploited by some foreign visitors, who steal “unpublished data, laboratory designs, grant proposals, experiment processes, research samples, blueprints, and state-of-the-art software and hardware” (DOJ, 2018b, p. 3). He also warned that visitors can exploit the open environments of these institutions, enabling them to spot talent and collect insights.²² Of particular concern, he said, is the use of foreign academics by their home countries’ intelligence services, which do not necessarily send or task academics with particular objectives, but rather seek to leverage them once they return home for a visit or upon the completion of their studies.

Of particular concern to some U.S. government officials are foreign talent recruitment programs, such as China’s Thousand Talents Program, through which foreign countries offer salaries, research facilities, and titles to induce expatriate scientists and other overseas experts to bring their knowledge and experience to China. China describes its Thousand Talents Program as a search for “strategic scientists or leading talents who can make breakthroughs in key technologies or can enhance China’s high-tech industries and emerging disciplines.”²³ The program seeks to recruit Chinese scholars currently living and working aboard, entrepreneurs, non-Chinese scholars, and younger scholars for long- and short-term appointments. U.S. officials characterize such programs as “compounding the threat” and encouraging the theft of IP (DOJ, 2018b, p. 4), and official presentations have described access to IP as these programs’ “key qualification” (NIH, 2018, chart 7).

Coincident with issuance of these warnings, NIH sent letters to more than 10,000 research institutions warning that “some foreign entities have mounted systematic programs … to take advantage of the long tradition

___________________

²² In addition to conducting illicit technology transfer, Priestap stated that foreign visitors exploiting access to U.S. institutions can introduce propaganda platforms, conduct training, recruit on behalf of foreign intelligence agencies, and stymie freedom of speech.

²³ “The Thousand Talents Plan,” 1000plan.org.cn/en/history.html.

of trust, fairness, and excellence of NIH-supported research activities.”²⁴ The letter highlighted three areas of concern, which it said were not limited to biomedical research but have long been posed as well by defense and energy research: diversion of IP; sharing of confidential information from grant proposals; and failure to disclose resources obtained

___________________

²⁴ Letter from NIH director Francis Collins, August 20, 2018, available at https://www.sciencemag.org/sites/default/files/NIH%20Foreign%20Influence%20Letter%20to%20Grantees%2008-20-18.pdf.

from other organizations, including foreign governments. The letter also invited research institutions to request briefings on these risks from FBI field offices. Concurrently, NIH privately reached out to grantee institutions with specific concerns; for example, NIH raised questions with the MD Anderson Cancer Center in Houston about three of its researchers, reported to be ethnic Chinese, who had reportedly failed to disclose foreign ties and had breached confidentiality (Tollefson, 2019; Zaveri, 2019). The Center moved to dismiss the three, two of whom chose to resign instead. And in early 2019, Emory University announced it had

fired two investigators who had failed to inform the university of their research affiliations with Chinese institutions (Tollefson, 2019). A similar notification from NIH was sent to Baylor College of Medicine regarding four faculty members. Rather than take steps to remove these faculty, the institution reviewed its policies and worked with the faculty to aid them in fully disclosing and describing their foreign collaborations (Ackerman, 2019). By June 2019, NIH had notified 61 institutions of apparent violations of rules concerning foreign relationships and had referred 16 cases to the U.S. Department of Health and Human Services’ Inspector General (Mervis, 2019c).

Although these actions have involved Chinese researchers, and NIH has acknowledged that China has been a significant focus of investigation, officials at NIH and affected institutions maintain that these actions are not motivated by, and do not constitute, racial profiling. According to a senior NIH official, “we’re focusing on objective behaviors. Not all of them involve China, and not all of the scientists whom we have discovered problems with are Chinese” (Tollefson, 2019). Nevertheless, the limited public detail behind these situations has given rise to concern among Chinese American and Chinese-origin researchers that the United States may not be a welcoming place for them (Tollefson, 2019).

Allegations by U.S. officials encompass several related issues that can be considered aspects of research misconduct, or the violation of academic norms or commitments: violations of the terms and conditions of federal grants that require disclosure of foreign financial conflicts and affiliations, unauthorized dissemination of proposals that have been circulated for confidential peer review, and theft of nonpublished research information (such as information obtained from the peer review of research manuscripts or through informal discussions).²⁵

Disclosure of foreign financial conflicts is important, as described by the head of NIH’s extramural research program, Michael Lauer, to prevent duplicative funding for the same research. Moreover, some Thousand Talents awards have required that IP developed in China remain in China and not be reported to U.S. institutions (Mervis, 2019b), conditions that might have affected NIH’s willingness to fund the work in the first place or that could have prompted it to attach further conditions to its support.

Similarly, violation of confidentiality in peer review is a clear-cut violation of academic practices. Grant proposals contain a scientist’s unique insights into how a problem can best be studied. They are circulated for review to experts in the same field who can understand the importance of the work and the feasibility of the proposed approach, and who conduct

___________________

²⁵ Two references describing norms of responsible research are NAS et al. (2009) and InterAcademy Partnership (2016).

these reviews confidentially to protect those insights from disclosure and possible application by competitors. Violating confidentiality is a violation of academic integrity regardless of who commits it, but such violations assume additional security and economic significance when they benefit scientists or economic interests in a competing nation. Additionally, some agencies, such as NIH, require that reviewers certify that they will not disclose grant information, and violations in those instances could therefore have legal ramifications.²⁶

Theft or unauthorized disclosure of nonpublic information or IP from a proprietary (typically corporate) research institution is similarly conceptually clear-cut. Such institutions seek competitive advantage through their research efforts, and disclosure of research results or methods enables competitors to benefit from the same information without having to bear any of the associated costs. In an academic setting, however, disclosure of nonpublic information, such as prepublication scientific results, is more complicated. The ultimate objective of most academic research is full and open publication, not only of the research results but also of the methods used to obtain them, and at a level of detail sufficient to allow any suitably trained and equipped researcher to duplicate (and hence validate) those results. Universities and other fundamental research institutions exist to generate and share information while training the next generation of researchers in the process. All of those who graduate from academic institutions, or who leave one laboratory or job to join or found another, do so with the expectation that they will bring the expertise they have acquired in their previous position to their new one. So while they all have the obligation to protect unpublished or confidential information and to respect IP, the idea that foreign researchers who come to American universities will not leave with any knowledge and technology is not well reasoned. Moreover, foreign researchers are often members of U.S.-funded scientific teams and contribute their intellectual capital to their projects’ success. Openness, engagement, and academic freedom have proven to be extremely effective in driving not just American scientific advances but American innovation—innovation that might have been stifled had the research been conducted under more restrictive conditions.

Even though the majority of unpublished research information is eventually disclosed, premature disclosure can give a head start to potential competitors, to the detriment of the originating laboratory. Moreover, some information associated with the research process may never be intended for publication. Again, violation of the academic obligation to respect the confidentiality of such information will harm American

___________________

²⁶ See https://grants.nih.gov/policy/research_integrity/confidentiality_peer_review.htm#prohibitions.

research, especially if it is being done—as is alleged by some U.S. officials—as a coordinated, systematic effort on the part of a competing nation (DOJ, 2018b; FBI, n.d.). If any individual is known to be violating these norms or is reasonably suspected of being likely to do so, action to mitigate that threat can be taken, whether by denying the individual an academic appointment, denying an appropriate visa, or removing the person from the research environment. Corrective action is much more difficult when a country may be suspected of fostering such activities among its nationals, but it is unclear who the specific offending parties may be. Blanket actions against—or scrutiny of—researchers solely on the basis of their nationality has the potential to degrade the openness that underlies the American research enterprise, and it can create the very “culture of suspicion” that MIT President Reif (2019) warned against.

Policies regarding foreign talent recruitment programs are challenged to find the appropriate balance. The first publicized U.S. government action against researchers engaged in foreign talent recruitment programs was taken by DOE in June 2019, when it issued a directive prohibiting DOE employees or contractors (including extramural researchers receiving DOE grants) from participating in the talent recruitment program of any country designated by DOE as a “foreign country of risk” (DOE, 2019). According to Under Secretary of Energy Paul Dabbar, “If you’re working for [DOE], and taking taxpayer dollars, we don’t want you to work for [foreign countries] at the same time” (Mervis, 2019a).

NIH appears to be taking a slightly different position. According to NIH Extramural Program Director Lauer, “Thousand Talents is not a threat [to the United States]…. It’s not the specific conduct we are focusing on, it’s the failure to disclose it” (Mervis, 2019b). It is not clear whether the apparent discrepancy between the policies of DOE and NIH is merely a matter of how each policy is described, represents differences in agency views that remain to be harmonized, or stems from the differences in the missions of the two agencies and their security cultures.

This committee is not in a position to evaluate all of the risks of foreign engagement, since not all the details in such cases as those mentioned here are publicly known. Moreover, some types of alleged improper behavior that might fall under the rubric “academic espionage,” when examined closely, appear to be an inherent consequence of openness, whereas others may require carefully balanced policy measures to address.

The committee does wish to acknowledge, as stated earlier in this chapter, that restrictions on foreign engagement at U.S. research institutions, even if deemed necessary, come at a price. Moreover, the perceptions generated by such actions can have serious consequences, particularly if not all of the underlying explanatory evidence can be made clear. Hence,

even if some direct harms can be attributed to inappropriate academic engagements such as those described here, the consequences of policy countermeasures may do more damage to the U.S. bioeconomy than the problem they are intended to solve. In congressional testimony on the importance of openness to U.S. education and research, former MIT President Charles Vest (2013) said he believed in the “leaky bucket theorem”: when it comes to research and technology, “it is far more important to keep filling our bucket than it is to obsessively plug leaks.”

In any event, any such policy instituted on the basis of a security perspective alone, without incorporating scientific and economic perspectives, risks being as one-sided as a policy instituted with no consideration of security at all. Given that science, economic, and security benefits are all at stake, a balanced policy process would involve all three.

State Involvement in Business Activities

An uneven international business landscape represents a substantial risk to the U.S. bioeconomy and puts U.S. companies at a disadvantage relative to some foreign competitors. For instance, the successful implementation of China’s Made in China 2025 plan to transform that country into a world leader in 10 high-tech sectors, including biomedicine and high-performance medical instruments, by 2025 has the potential to disadvantage U.S. companies relative to their Chinese counterparts. According to the FBI, China plans to eliminate all foreign-produced technology in these sectors by 2025. A public document prepared by the FBI to educate the academic sector about the potential risks to academia states, “The Chinese government uses numerous methods—some legitimate but others, such as stealing technology from foreign competitors, meant to illicitly introduce foreign technology and knowledge to China” (FBI, n.d., p. 3). There are also a number of other reports and studies pointing to similar concerns (Brown et al., 2018; Morrison, 2019; U.S. Chamber of Commerce, 2017). According to the Office of the United States Trade Representative (USTR, 2018a), a state-directed economic program provides government subsidies for Chinese companies and mobilizes state-backed financial institutions to fund the acquisition of foreign biotech companies, with the goal of acquiring IP, and artificially distorts the market to establish Chinese companies as world leaders. An example is the $43 billion acquisition of Syngenta by the China National Chemical Corp. (ChemChina), a state-owned Chinese chemical company (Shields, 2017). The acquisition included Syngenta’s entire U.S. business of more than 4,000 employees, 33 research sites, and 31 production and supply sites. The transaction was financed in part by a consortium of state-run financial entities. Critics

argue that state-directed investment on this scale undermines the principles of open trade and distorts global markets, prioritizing political considerations to the detriment of scientific innovation and normal economic incentives.

Under the 1988 Exon-Florio Amendment, as amended by the Foreign Investment Risk Review Modernization Act of 2018, the President has the power to block investments by foreign entities in U.S. companies or real estate when those investments may impair U.S. national security—for example, by putting technologies, data, or capabilities relevant to national security under foreign control.²⁷ In practice, the cabinet-level Committee on Foreign Investment in the United States (CFIUS) will try to work with parties to a transaction to mitigate any risk to national security. However, if the parties to the proposed transaction cannot reach an agreement on mitigation measures that satisfy the Committee, the Committee can recommend that the President block the transaction in its entirety. (The President also has the ability to reverse those types of transactions if they occurred without review and approval.) CFIUS was recently given extended authority to review transactions involving not just foreign ownership but also other investments that might afford foreign persons access to nonpublic technical information in the possession of certain U.S. businesses, along with any other “transaction, transfer, agreement, or arrangement designed to circumvent CFIUS.”²⁸ Foreign investment controls may impose an economic price on particular firms by precluding them from accessing certain foreign sources of investment, but in the longer run they may advantage U.S. firms by slowing or preventing the loss of information or technology that can be used by foreign competitors.

Trade Barriers

The U.S. bioeconomy, like other aspects of the U.S. economy, relies on fair access to domestic and international markets for dissemination of products and services. Therefore, asymmetries in trade practices, such as regulatory approval processes for foreign products and forced

___________________

²⁷ On the U.S. Department of the Treasury website, see “Section 721 of the Defense Production Act of 1950, 50 USC App. 2170 (as amended by the Foreign Investment and National Security Act of 2007),” https://www.treasury.gov/resource-center/international/foreign-investment/Documents/Section-721-Amend.pdf; and “Summary of the Foreign Investment Risk Review Modernization Act of 2018,” https://www.treasury.gov/resourcecenter/international/Documents/Summary-of-FIRRMA.pdf.

²⁸ See https://www.treasury.gov/resource-center/international/Documents/Summary-ofFIRRMA.pdf.

technology transfer practices, have the potential to hinder or harm the U.S. bioeconomy.

Asymmetric Regulatory Practices

Asymmetric regulatory practices between trading partners have the potential to affect the ability of domestic companies to reach foreign markets. With respect to the bioeconomy, this is particularly the case for agricultural biotechnology and the pharmaceutical sector. If applicable regulations for a given product are not harmonized among major global markets, innovations from one nation will have difficulty gaining full or timely reach into the global bioeconomy. And when different countries or trade blocks take philosophically different approaches to regulation, as do the United States, with its largely product-based regulatory system, and the European Union, with its more process-based system, the problem is not just that products will obtain different regulatory approvals at different times in different jurisdictions, but that products regulated in one jurisdiction may be completely unregulated in another.

The United States generally aspires to regulate new crops improved through biotechnology under a risk-based, science-based framework that treats products according to the risks they pose, independent of the process by which they were generated. The European Union takes a precautionary approach in which genetically modified crops must undergo risk analyses not required for unmodified crops. In 2003, the United States filed a complaint with the World Trade Organization (WTO) claiming that the European Union’s de facto moratorium on the approval of genetically modified crop imports violated WTO agreements (Chereau, 2014). In 2006, WTO ruled that this moratorium and the genetically modified organism (GMO) approval processes of several European Union states were illegal. In 2013, the European Union General Court ruled that the European Union must process a long-pending authorization to import a genetically modified corn (Law Library of Congress, 2014). However, a number of European states continue to oppose the decision.

The United States considers the European Union’s GMO approval policies to be inconsistent with a risk-based approach to the regulation of agriculture, and it regards the WTO ruling as confirmation that these policies constitute an unwarranted barrier to trade. The United States views the policies not only as denying access to markets in the European Union but also denying U.S. companies access to markets in other countries (outside of the European Union) that fear they will not be able to export the resulting crops to the European Union.

Another example within the agricultural sector, although with very different implications, relates to the more permissive regulatory environment for gene-edited livestock. In 2008, FDA issued guidance stating that it would regulate genetically engineered animals under the Federal Food, Drug, and Cosmetic Act. FDA considered the use of the recombinant DNA used to create the genetic modification to represent a “new animal drug,” and therefore subject to “government review and approval, the same as a veterinary drug such as an antibiotic or pain reliever” (Miller and Cohrssen, 2018). This guidance and the subsequent requirements for labeling led to an 11-year-long regulatory review and labeling decision-making process for a genetically engineered salmon (Clayton, 2019). As a result, other American companies and researchers working on gene editing of other animal species for food (such as hornless cattle, heat-resistant cattle, goats with an antimicrobial protein in their milk, and disease-resistant pigs) have decided to move their research and production to Argentina, Australia, Brazil, and Canada (Ledford, 2019). In short, the implications of the slow and uncertain regulatory process are causing some American companies to move overseas, thus potentially leaving the United States behind.

There is also a lack of reciprocity with respect to pharmaceutical licensing. The regulatory rules for approval of pharmaceuticals in China are opaque, and decisions to approve are based on factors other than science. For example, if a U.S. drug company wants to license a drug in China, it must complete the approval process in the United States before it can begin the approval process in China, whereas other countries allow for concurrent clinical trials. This practice limits the time that a U.S. drug company can market a patented drug in China.

Forced Technology Transfer

China’s noncompliance with some international business norms and WTO rules, particularly with respect to forced technology transfer, have been documented by USTR (2018b). According to the USTR report, the Chinese government forces the transfer of foreign companies’ technologies and IP to Chinese companies though opaque administrative licensing and approval processes, noting that “Chinese officials may use oral communication and administrative guidance to pressure foreign firms to transfer technology.” Such policies clearly disadvantage U.S. firms relative to Chinese firms, which face no such barrier selling products in the United States. The USTR’s 2018 report to Congress on China’s WTO compliance, issued in February 2019, observes that, “despite repeated commitments to refrain from forcible technology transfer from U.S. companies,

China continues to do so through market access restrictions, the abuse of administrative processes, licensing regulations, asset purchases, cyber and physical theft” (USTR, 2019).

The Bioeconomy as a Component of Critical Infrastructure

In the United States, critical infrastructures include the financial sector, the electrical power grid, transportation systems, energy systems, communications systems, and a range of others.²⁹ To the extent that the bioeconomy becomes increasingly integrated into these critical infrastructures, incapacitation or failure of key bioeconomy facilities or services could also threaten security, public health, or public safety. For example, the production of vaccines for public health could be considered part of the critical health care and public health infrastructure. Foods, fuels, and medicine can all be considered critical to the nation’s health and stability. Therefore, to the extent that they are produced on automated and digital biotech platforms, their cyber and other vulnerabilities need to be recognized and specifically addressed.

Incapacitation of critical facilities need not, however, require a natural disaster or a physical or cyberattack. The operation of any bioeconomy facility that is dependent on input materials available only overseas is subject to interruption if the supply chains for those inputs are interrupted, whether by decisions of foreign powers to withhold shipment or by failures of international transportation networks.³⁰ Moreover, dependence on imports exposes the United States to potential sources of counterfeit or adulterated products if the regime ensuring product integrity in the supplier country is inadequate.³¹ Protecting against such interruptions requires developing multiple secure sources of supply for critical inputs, stockpiling the inputs, or engineering around these dependencies.

___________________

²⁹ Presidential Policy Directive 21, “Critical Infrastructure Security and Resilience” (White House Office of the Press Secretary, 2013), see footnote 23.

³⁰ Note that “supply chain” in this discussion refers to the routes by which the materials and components necessary to produce some output of the bioeconomy are integrated into final products or used in delivering services. This meaning is different from the phrase’s use in “supply chain attacks,” as discussed in the section on cybersecurity, which refers to engineering flaws into component systems with the expectation that those components would be incorporated in more complicated systems that could then be penetrated by exploiting those flaws.

³¹ In 2008, for example, 81 deaths were associated with contaminated supplies of the blood thinner heparin produced by 12 Chinese companies and exported to 11 countries. The companies apparently all drew on supplies of an active ingredient contaminated with a chemical that was difficult to distinguish from heparin but was much cheaper (Greenemeier, 2008; Harris, 2008; Powell, 2008).

Traditional Biosecurity and Biosafety Risks

The tools of today’s bioeconomy are enabling new capabilities that can generate concerns regarding traditional biothreats, which encompass primarily those pathogens considered to be most dangerous and lethal or used as weapons in the past. These agents were placed on security lists, such as the Federal Select Agent List, to protect against their unauthorized acquisition, possession, and use.³² Cold War–era bioweaponeers wanted to alter pathogens to make them deadlier, to spread more easily, or able to evade diagnosis and treatment, but these goals required heavy investment, expertise, and time commitment and faced knowledge and technical barriers. With today’s tools, however, the acquisition of dangerous pathogenic organisms can be facilitated through synthetic creation “from scratch” based on their known genomic sequences, and DNA is commercially available from a growing number of gene synthesis companies throughout the world. Recent examples include reconstruction of such viruses as polio (Cello et al., 2002), the 1918 influenza virus (Tumpey et al., 2005), and most recently horsepox (Noyce et al., 2018). Such developments as the efficient genome editor CRISPR illustrate the programmable tools that could rewrite genetic code to alter pathogens in ways aspired to by weapons programs of the past. Another possibility is the creation of novel bioweapons that do not currently exist on any security control lists and would be difficult to prevent, detect, and treat. A 2018 study by the National Academies highlights the most concerning capabilities stemming from synthetic biology that could harm humans (NASEM, 2018a).

Although manipulation of pathogenic organisms remains technically challenging, then, the tools of today’s biotechnology could lower the technological barriers (DiEuliis, 2019). A strong bioeconomy will also pursue the ability to manipulate biological organisms, the same capabilities that could drive bioweapons programs, albeit for different intentions and with different organisms. It remains to be seen whether expansion of the knowledge base and specialized tools for bioeconomic products, such as the open-source biology movement, can also serve in similar kinds of manipulations of pathogens (Cohn, 2005). Presumably, these capabilities will require tailored bioinformatics, which may also be applicable to making or tinkering with harmful pathogens if made broadly available (i.e., not protected as IP within companies).

___________________

³² The Federal Select Agent Program regulates the possession, transport, and use of certain biological pathogens that are considered to pose a severe threat to public, animal, or plant health or safety. This program and the agents it regulates are described at www.selectagents.gov.

It is important to note that, while growing pathogens to scale, storing them stably, and delivering them to target populations are the most challenging aspects of bioweapon development, some of these capabilities are real and purposeful goals of the bioeconomy—to scale up production of organisms that can produce high-value products or themselves be used as products. As industry continues to resolve challenges in the creation of chemicals, there is also a growing overlap between biological and chemical weapons. Importantly, biothreats to humans are only one component of the risk; threats to animals, plants, agriculture, the environment, and materials are also of concern. While these potential enablers of biothreats cannot and should not be minimized, strong public health and animal health infrastructures will still serve as robust primary defenses.

Certain U.S. export controls serve as one means of countering some biosecurity concerns. For national security or foreign policy purposes, the U.S. government requires that licenses be obtained for the export of some goods, technology, and information to certain destinations to prevent their falling into the hands of adversaries. Moreover, communication of controlled technical information within the United States to a foreign national is deemed an export to that individual’s country, and may require an export license as well. Fundamental research—defined broadly as research intended to be openly published—is not subject to export controls, but such controls may apply to information that is protected as proprietary or is otherwise not public.

Export controls may have the effect of preventing entities abroad from acquiring technology that could allow them to compete with U.S firms. It is important to note, however, that these controls can serve to protect national security at the expense of competitiveness, because U.S. firms may be precluded from selling products to certain foreign customers, and foreign manufacturers may have an incentive to avoid the use of U.S. components to prevent triggering the imposition of U.S. export controls.

Almost all bioeconomy-related items that are subject to export controls fall under the Export Administration Regulations (EAR), which establish export controls on so-called “dual-use” items, which as noted earlier are commercial items that can also be used for military or terrorist purposes.³³ The U.S. Department of Commerce’s Bureau of Industry and Security administers the EAR, which include the Commerce Control List that

___________________

³³ The term “dual-use” also describes research done for legitimate purposes that can be misused for harm, but that definition is not relevant to export controls.

describes items subject to dual-use controls.³⁴ Because controls on goods by one country can be undercut if other exporting countries do not control the same items, nations work together to coordinate and harmonize their export control systems. Controls on items related to chemical and biological weapons are coordinated informally (i.e., in the absence of a formal mechanism such as a treaty) through the Australia Group. Australia Group members meet periodically to consider changes to the list of controlled items. The United States also has the ability to control items unilaterally.

Under a provision in the National Defense Authorization Act of 2019, Congress called for the U.S. Department of Commerce to establish export controls on emerging and foundational technologies that are “essential to the national security of the United States.” This process is intended to take into account the status of development of these technologies in foreign countries, the effect such controls might have on their development in the United States, and the potential effectiveness of the controls in curbing the proliferation of these technologies.³⁵ On November 19, 2018, the U.S. Department of Commerce issued an Advance Notice of Proposed Rule-making (ANPRM) in the Federal Register to solicit public comment on how emerging technologies could be identified and assessed for the purpose of updating export control lists.³⁶ The ANPRM asked in particular about whether biotechnology should be considered for controls, and also reiterated that the Department does not seek to expand export controls into areas not currently subject to them, such as fundamental research.

Even with this qualification, many respondents to the ANPRM warned that instituting controls not precisely targeting specific technological developments would harm the United States’ ability to develop emerging technologies. A consortium of academic organizations warned that “overly broad or vague controls will result in unnecessary regulations that will stifle scientific progress and impede research.”³⁷ The Biotechnology Innovation Organization cautioned the Department to “move with extreme caution to avoid unintended harm to U.S. domestic research and development of novel biotechnologies, U.S. international competitiveness,

___________________

³⁴ A separate system of export controls, run by the U.S. Department of State, governs the export of weapons systems and military-specific technologies. This system, administered as the International Trafficking in Arms Regulations, is less relevant to the bioeconomy.

³⁵ National Defense Authorization Act of 2018, P.L. 115-232, § 1758(a)(1).

³⁶U.S. Department of Commerce, 2018. Foundational technologies are to be addressed in a subsequent ANPRM.

³⁷ Letter to the U.S. Department of Commerce from the Council on Governmental Relations, the Association of American Universities, the Association of Public and Land-grant Universities, the American Council on Education, and the AAMC (formerly the Association of American Medical Colleges), January 10, 2019, available at https://www.regulations.gov/document?D=BIS-2018-0024-0140.

and economic growth,” pointing out that the biotechnology industry is an inherently global ecosystem and utilizes global clinical research partnerships.³⁸ The U.S. Department of Commerce received 247 responses to its request for comment and as of this writing had not responded to them.

Risks from Global Climate Change

Global climate change will significantly affect the bioeconomy even as the bioeconomy provides means to help offset greenhouse gas emissions by providing a biobased pathway for the creation of products that are currently dependent on fossil fuels (such as petroleum-based plastics). Food and feed crops, lignocellulosic bioenergy crops, and crops grown for plant-derived sugars as feedstock for fermentative processing are susceptible to temperature and water stresses, and they will be vulnerable to insects and pathogens that migrate from their current habitats. The government’s forecast on the impacts of climate change on agriculture states that the largest contributing factor to declines in U.S. agricultural productivity will be increases in temperature during the growing season in the Midwest (USGCRP, 2018). Arresting climate change–induced declines in agricultural productivity will require improvements in three dimensions—quality, yield, and an optimized and sustainable system that does not compromise benefits of the system. Moreover, while some crops, such as grain and biomass sorghum, may be able to withstand climate change–induced stresses such as drought, for most species, mitigation will require identifying more resilient genotypes from among naturally occurring diversity, engineering them for greater resilience, or moving crop cultivation to areas that replicate the climate in which they are currently grown (which has obvious geographic and land-use implications).

While global climate change is an existential threat that specifically affects agricultural production at the foundation of the bioeconomy, partial mitigation can be accomplished through long-term and strategic support of a vibrant bioeconomy as discussed in the recommendations in this report.

CONCLUSIONS

This chapter reviewed the risks identified by the committee that have the potential to adversely affect the U.S. bioeconomy. Where possible, the committee has discussed some of the policy tools that can be used to

___________________

³⁸ Letter to the U.S. Department of Commerce from the Biotechnology Innovation Organization, January 10, 2019, available at https://www.regulations.gov/document?D=BIS-2018-0024-0137.

mitigate these risks. It is important to recognize that some of the identified policy actions have the potential to cause unintended consequences or outcomes. This potential is best illustrated by, but is not limited to, the concerns around foreign researchers or the regulatory system. Over the course of its deliberations, the committee arrived at a number of conclusions related to the risks facing to the U.S. bioeconomy.

Conclusion 7-1: Limitations on fundamental research, whether through a lack of support, the implementation of restrictive research regulations, or the inability to develop and attract a skilled workforce, could erode the United States’ ability to produce breakthrough scientific results and develop enabling technologies.

Conclusion 7-2: Access to data is vital to the bioeconomy research enterprise, and issues related to data sharing (domestically or internationally), benefit sharing, or the potential use of data for malicious reasons will require carefully considered solutions.

Conclusion 7-3: The bioeconomy faces many of the traditional cybersecurity risks faced by other sectors. Common features of the bioeconomy that pose potential vulnerabilities include reliance on open-source software, large and potentially sensitive datasets, and communication through the Internet (such as via networked devices that are potentially running outdated software).

Conclusion 7-4: Concerns about foreign researchers, potential policy actions to address those concerns, and the perceptions generated by such actions have the potential to adversely affect the bioeconomy if not informed by input from the scientific community. Given that science, economic, and security benefits are all at stake, a balanced policy process would involve all three perspectives.

Conclusion 7-5: More information is needed to understand the impact of current and proposed requirements for patent eligibility on the sustainability and growth of the U.S. bioeconomy. Specifically, more information is needed regarding the extent to which patent eligibility requirements impact the ability of start-up companies and larger, well-established companies to secure patent protection in the United States, and whether these companies are more or less inclined toward or successful in securing patent protection internationally.

Conclusion 7-6: International asymmetries regarding the regulation of bioeconomy products, data-sharing agreements and practices,

and industrial mergers and acquisitions (including associated technology transfers and potential state involvement) are risks to the U.S. bioeconomy.

The discussion of risks and potential policy responses in this chapter has stressed the importance of finding the right balance between protecting the U.S. research enterprise and the safety of bioeconomy products, on the one hand, and not unduly impeding innovation in and the growth of the bioeconomy on the other. This issue is addressed further in the next chapter, which presents the committee’s overarching conclusions and recommendations.

REFERENCES

Bahr, R. W. 2018b. Recent subject matter eligibility decision: Vanda Pharmaceuticals Inc. v. WestWard Pharmaceuticals. Alexandria, VA: U.S. Patent and Trademark Office. https://www.uspto.gov/sites/default/files/documents/memo-vanda-20180607.PDF (accessed October 19, 2019).

Bailey, M. N., and N. Montalbano. 2017. Clusters and innovation districts: Lessons from the United States experience. Washington, DC: Brookings Institution. https://www.brookings.edu/wp-content/uploads/2017/12/es_20171208_bailyclustersandinnovation.pdf (accessed September 5, 2019).

Bavasi, H., L. Thornton, D. Peloquin, and M. Barnes. 2017. Using biospecimens collected abroad in future research: Key considerations. Medical Research Law & Policy Report. U.S. Bureau of National Affairs. https://www.ropesgray.com/~/media/Files/articles/2017/April/Bloomberg%20BNA%20Biospecimens%20Abroad%20for%20Research%20Article%204-19-17.ashx (accessed May 23, 2019).

Black, P. E., L. Badger, B. Guttman, and E. Fong. 2016. Dramatically reducing software vulnerabilities: Report to the White House Office of Science and Technology Policy. NISTIR 8151. https://nvlpubs.nist.gov/nistpubs/ir/2016/NIST.IR.8151.pdf (accessed September 5, 2019).

Braudt, D. B. 2018. Sociogenomics in the 21st century: An introduction to the history and potential of genetically-informed social science. Sociology Compass 12(10):e12626. doi: 10.1111/soc4.12626.

Broniatowski, D. A., A. M. Jamison, S. Qi, L. AlKulaib, T. Chen, A. Benton, S. C. Quinn, and M. Dredze. 2018. Weaponized health communication: Twitter bots and Russian trolls amplify the vaccine debate. American Journal of Public Health 108(10):1378–1384. doi: 10.2105/AJPH.2018.304567.

Brown, M., and P. Singh. 2018. China’s technology transfer strategy. https://admin.govexec.com/media/diux_chinatechnologytransferstudy_jan_2018_(1).pdf (accessed October 16, 2019).

Cancer Genome Atlas Research Network. 2013. The cancer genome atlas pan-cancer analysis project. Nature Genetics 45:1113. doi: 10.1038/ng.2764.

CDC and USDA (U.S. Centers for Disease Control and Prevention and U.S. Department of Agriculture). 2017a. Information systems security controls guidance. https://www.selectagents.gov/resources/Information_Systems_Guidance.pdf (accessed September 5, 2019).

CDC and USDA. 2017b. Security plan guidance. https://www.selectagents.gov/resources/Security_Plan_Guidance.pdf (accessed September 5, 2019).

Cello, J., A. V. Paul, and E. Wimmer. 2002. Chemical synthesis of poliovirus cDNA: Generation of infectious virus in the absence of natural template. Science 297(5583):1016–1018.

Chao, B., and A. Mapes. 2016. An early look at Mayo’s impact on personalized medicine. Patently-O Patent Law Journal 10:10–14.

Chereau, C. J. 2014. Biotechnology: Can the transatlantic trade and investment partnership reconcile EU and U.S. differences on GMOs? International Journal of Trade and Global Markets 7(4):316–338.

Clayton, C. 2019. AquaBounty salmon no longer swimming up regulatory stream. Ag Policy Blog, July 11. https://www.dtnpf.com/agriculture/web/ag/blogs/ag-policy-blog/blog-post/2019/07/11/aquabounty-salmon-longer-swimming-2 (accessed October 19, 2019).

Cohen, J. 2018. A chat with the geneticist who predicted how the police may have tracked down the Golden State Killer. Science News, April 29. https://www.sciencemag.org/news/2018/04/chat-geneticist-who-predicted-how-police-may-have-tracked-downgolden-state-killer (accessed October 16, 2019).

Cohn, D. 2005. Open-source biology evolves. Wired, January 17. https://www.wired.com/2005/01/open-source-biology-evolves (accessed September 5, 2019).

Comfort, N. 2018. Sociogenomics is opening a new door to eugenics. MIT Technology Review, October 23. https://www.technologyreview.com/s/612275/sociogenomics-is-opening-a-new-door-to-eugenics (accessed October 16, 2019).

Council of Economic Advisors. 2018. The cost of malicious cyber activity to the U.S. economy. https://www.whitehouse.gov/wp-content/uploads/2018/03/The-Cost-of-MaliciousCyber-Activity-to-the-U.S.-Economy.pdf (accessed August 9, 2019).

CSIS (Center for Strategic and International Studies). 2005. Security controls on the access of foreign scientists and engineers to the United States. https://www.csis.org/analysis/security-controls-access-foreign-scientists-and-engineers-united-states (accessed June 9, 2019).

Dankar, F. K., A. Ptitsyn, and S. K. Dankar. 2018. The development of large-scale de-identified biomedical databases in the age of genomics—principles and challenges. Human Genomics 12(1):19. doi: 10.1186/s40246-018-0147-5.

Dhar, D., and J. H.-E. Ho. 2009. Stem cell research policies around the world. Yale Journal of Biology and Medicine 82(3):113–115.

DiEuliis, D. 2018. Biotechnology for the battlefield: In need of a strategy. War on the rocks, November 27. https://warontherocks.com/2018/11/biotechnology-for-the-battlefield-in-need-of-a-strategy (accessed July 2, 2019).

DiEuliis, D. 2019. Key national security questions for the future of synthetic biology. Fletcher Forum of World Affairs 43(1):1.

DOE (U.S. Department of Energy). 2019. Department of Energy foreign government talent recruitment programs. DOE Order O 486.1. https://www.energy.gov/sites/prod/files/2019/06/f63/DOE%20O%20486.1.pdf (accessed June 18, 2019).

DOJ (U.S. Department of Justice). 2018a. Chinese scientist sentenced to prison in theft of engineered rice. https://www.justice.gov/opa/pr/chinese-scientist-sentenced-prison-theft-engineered-rice (accessed September 6, 2019).

DOJ. 2018b. Statement of E. W. Priestap, Assistant Director, Counterintelligence Division, Federal Bureau of Investigation, before the Committee on the Judiciary, Subcommittee on Border Security and Immigration, U.S. Senate, at a hearing “Student Visa Integrity: Protecting Educational Opportunity and National Security,” June 6. https://www.judiciary.senate.gov/imo/media/doc/06-06-18%20Priestap%20Testimony.pdf (accessed October 16, 2019).

DOJ. 2019. Interim policy: Forensic genetic genealogical DNA analysis and searching. https://www.justice.gov/olp/page/file/1204386/download (accessed October 16, 2019).

DOJ Office of Public Affairs. 2019. Department of Justice announces interim policy on emerging method to generate leads for unsolved violent crimes. Press release 19-1017. https://www.justice.gov/opa/pr/department-justice-announces-interim-policy-emerging-methodgenerate-leads-unsolved-violent (accessed October 16, 2019).

EBRC (Engineering Biology Research Consortium). 2019. Engineering biology: A research roadmap for the next-generation bioeconomy. https://roadmap.ebrc.org (accessed July 3, 2019).

Ellis, S. 2018. U.S. biopharma firms hit by cyber attacks from China. Bioworld. http://www.bioworld.com/content/us-biopharma-firms-hit-cyber-attacks-china-0 (accessed November 9, 2019).

Erlich, Y., T. Shor, I. Pe’er, and S. Carmi. 2018. Identity inference of genomic data using long-range familial searches. Science 362(6415):690–694.

FBI (Federal Bureau of Investigation). n.d. China: The risk to academia. https://www.research.psu.edu/sites/default/files/FBI_Risks_To_Academia.pdf (accessed August 12, 2019).

Feldman, M. P., and N. Massard, eds. 2002. Institutions and systems in the geography of innovation. New York: Kluwer Academic Publishers.

Fiume, M., M. Cupak S. Keenan, J. Rambla, S. de la Torre, S. O. M. Dyke, A. J. Brookes, K. Carey, D. Lloyd, P. Goodhand, M. Haeussler, M. Baudis, H. Stockinger, L. Dolman, I. Lappalainen, J. L. Törnroos, J. D. Spalding, S. Ur-Rehman, A. Page, P. Flicek, S. Sherry, D. Haussler, S. Varma, G. Saunders, and S. Scollen. 2019. Federated discovery and sharing of genomic data using Beacons. Nature Biotechnology 37(3):220–224.

Flach, J., C. S. Ribeiro, M. B. van der Waal, R. X. van der Waal, E. Claassen, L. H. M. van de Burgwal. 2019. The Nagoya protocol on access to genetic resources and benefit sharing: Best practices for users of lactic acid bacteria. PharmaNutrition 9:100158. doi: 10.1016/j. phanu.2019.100158.

Fredrikson, M., E. Lantz, S. Jha, S. Lin, D. Page, and T. Ristenpart. 2014. Privacy in pharmacogenetics: An end-to-end case study of personalized warfarin dosing. San Diego, CA: USENIX Association. https://www.usenix.org/system/files/conference/usenixsecurity14/sec14-paper-fredrikson-privacy.pdf (accessed October 19, 2019).

Galdzicki, M., C. Rodriguez, D. Chandran, H. M. Sauro, and J. H. Gennari. 2011. Standard biological parts knowledgebase. PLoS One 6(2):e17005. doi: 10.1371/journal. pone.0017005.

Greenemeier, L. 2008. Heparin scare: Deaths from tainted blood-thinner spur race for safe replacement. Scientific American, November 4. https://www.scientificamerican.com/article/heparin-scare-deaths (accessed September 5, 2019).

Gryphon Scientific and Rhodium Group. 2019. China’s biotechnology development: The role of U.S. and other foreign engagement: A report prepared for the U.S.-China Economic and Security Review Commission. https://www.uscc.gov/sites/default/files/Research/USChina%20Biotech%20Report.pdf (accessed July 1, 2019).

Guhr, A., S. Kobold, S. Seltmann, A. E. M. Seiler Wulczyn, A. Kurtz, and P. Löser. 2018. Recent trends in research with human pluripotent stem cells: Impact of research and use of cell lines in experimental research and clinical trials. Stem Cell Reports 11(2):485–496. doi: 10.1016/j.stemcr.2018.06.012.

Harris, G. 2008. U.S. identifies tainted heparin in 11 countries. The New York Times, April 22. https://www.nytimes.com/2008/04/22/health/policy/22fda.html (accessed July 5, 2019).

Hiemstra, S. J., M. Brink, and T. van Hintum. 2019. Digital Sequence Information (DSI): Options and impact of regulating access and benefit sharing—stakeholder perspectives. CGN Report 42. Wageningen, Netherlands: Wageningen University & Research. https://library.wur.nl/WebQuery/wurpubs/fulltext/470286 (accessed September 5, 2019).

Hiscox. 2018. Small Business Cyber Risk Report^TM. https://www.hiscox.com/documents/2018Hiscox-Small-Business-Cyber-Risk-Report.pdf (accessed October 16, 2019).

Hodgson, J. 2019. Genentech trade secrets theft highlights risk for biotechs. Nature Biotechnology 37(1):9–10.

Hyde, C. L., M. W. Nagle, C. Tian, X. Chen, S. A. Paciga, J. R. Wendland, J. Y. Tung, D. A. Hinds, R. H. Perlis, and A. R. Winslow. 2016. Identification of 15 genetic loci associated with risk of major depression in individuals of European descent. Nature Genetics 48(9):1031–1036. doi: 10.1038/ng.3623.

Ilic, D., and C. Ogilvie. 2017. Concise review: Human embryonic stem cells—What have we done? What are we doing? Where are we going? Stem Cells 35(1):17–25. doi: 10.1002/stem.2450.

InterAcademy Partnership. 2016. Doing global science: A guide to responsible conduct in the global research enterprise. Princeton, NJ: Princeton University Press. https://www.interacademies.org/33345/Doing-Global-Science-A-Guide-to-Responsible-Conduct-in-the-Global-Research-Enterprise (accessed October 19, 2019).

IOM (Institute of Medicine). 2012. Adverse effects of vaccines: Evidence and causality. Washington, DC: The National Academies Press. https://doi.org/10.17226/13164.

Jefferson, O. A., D. Köllhofer, T. H. Ehrich, and R. A. Jefferson. 2015. Gene patent practice across plant and human genomes. Nature Biotechnology 33(10):1033–1038.

Jouvenal, J. 2018. To find alleged Golden State Killer, investigators first found his great-great-great-grandparents. The Washington Post, April 30. https://www.washingtonpost.com/local/public-safety/to-find-alleged-golden-state-killer-investigators-first-found-hisgreat-great-great-grandparents/2018/04/30/3c865fe7-dfcc-4a0e-b6b2-0bec548d501f_story.html (accessed October 16, 2019).

Kaiser, J. 2018. We will find you: DNA search used to nab Golden State Killer can home in on about 60% of white Americans. Science News, October 11. https://www.sciencemag.org/news/2018/10/we-will-find-you-dna-search-used-nab-golden-state-killer-canhome-about-60-white (accessed October 15, 2019).

Laestadius, L. I., J. R. Rich, and P. L. Auer. 2017. All your data (effectively) belong to us: Data practices among direct-to-consumer genetic testing firms. Genetics in Medicine 19(5):513–520.

Lai, H.-E., C. Canavan, L. Cameron, S. Moore, M. Danchenko, T. Kuiken, Z. Sekeyová, and P. S. Freemont. 2019. Synthetic biology and the United Nations. Trends in Biotechnology 37(11):1146–1151. https://www.sciencedirect.com/science/article/pii/S0167779919301337 (accessed September 5, 2019).

Law Library of Congress. 2014. Restrictions on genetically modified organisms. https://www.loc.gov/law/help/restrictions-on-gmos/restrictions-on-gmos.pdf (accessed June 17, 2019).

Le Feuvre, R. A., and N. S. Scrutton. 2018. A living foundry for synthetic biological materials: A synthetic biology roadmap to new advanced materials. Synthetic and Systems Biotechnology 3(2):105–112. doi: 10.1016/j.synbio.2018.04.002.

Ledford, H. 2019. Gene-edited animal creators look beyond U.S. market. Nature 566:433–434. doi: 10.1038/d41586-019-00600-4.

Levine, D. S., and T. Sichelman. 2018. Why do startups use trade secrets? Notre Dame Law Review 94(2):751–820.

Majumder, M. A. 2018. United States: Law and policy concerning transfer of genomic data to third countries. Human Genetics 137(8):647–655. doi: 10.1007/s00439-018-1917-9.

Mantle, J. L., J. Rammohan, E. F. Romantseva, J. T. Welch, L. R. Kauffman, J. McCarthy, J. Schiel, J. C. Baker, E. A. Strychalski, K. C. Rogers, and K. H. Lee. 2019. Cyberbiosecurity for biopharmaceutical products. Frontiers in Bioengineering and Biotechnology 7:116. doi: 10.3389/fbioe.2019.00116.

Martin, C. 2018. Privacy and consumer genetic testing don’t always mix. Science News, June 5. https://www.sciencenews.org/blog/science-public/privacy-and-consumer-genetic-testing-dont-always-mix (accessed October 16, 2019).

Mervis, J. 2019a. DOE to limit foreign research collaborations. Science 363(6428):675–676.

Mervis, J. 2019b. NIH probe of foreign ties has led to undisclosed firings—and refunds from institutions. Science, June 26. https://www.sciencemag.org/news/2019/06/nih-probe-foreign-ties-has-led-undisclosed-firings-and-refunds-institutions (accessed August 12, 2019).

Mervis, J. 2019c. Powerful U.S. senator calls for vetting NIH grantees at hearing on foreign influences. Science, June 6. https://www.sciencemag.org/news/2019/06/powerful-us-senator-calls-vetting-nih-grantees-hearing-foreign-influences (accessed August 12, 2019).

Miller, H. I., and J. J. Cohrssen. 2018. FDA overreach on genetically engineered animals. Issues in Science and Technology 34(3).

Millett, P., T. Binz, S. W. Evans, T. Kuiken, K. Oye, M. J. Palmer. C. van der Vlugt, K. Yambao, and S. Yu. 2019. Developing a comprehensive, adaptive, and international biosafety and biosecurity program for advanced biotechnology: The iGEM experience. Applied Biosafety 24(2):64–71. doi: 10.1177/1535676019838075.

Monteiro, R. L. 2018. The new Brazilian General Data Protection Law—a detailed analysis. IAPP, August 15. https://iapp.org/news/a/the-new-brazilian-general-data-protection-law-a-detailed-analysis (accessed September 6, 2019).

Morrison, W. 2019. Enforcing U.S. trade laws: Section 301 and China. Congressional Research Service “In Focus” IF10708, updated June 26, 2019. https://fas.org/sgp/crs/row/IF10708.pdf (accessed October 1, 2019).

Murch, R. S., W. K. So, W. G. Buchholz, S. Raman, and J. Peccoud. 2018. Cyberbiosecurity: An emerging new discipline to help safeguard the bioeconomy. Frontiers in Bioengineering and Biotechnology 6:39. doi: 10.3389/fbioe.2018.00039.

NAS, NAE, and IOM (National Academy of Sciences, National Academy of Engineering, and Institute of Medicine). 2007. Rising above the gathering storm: Energizing and employing America for a brighter economic future. Washington, DC: The National Academies Press. https://doi.org/10.17226/11463.

NAS, NAE, and IOM. 2009. On being a scientist: A guide to responsible conduct in research, 3rd edition. Washington, DC: The National Academies Press. https://doi.org/10.17226/12192.

NAS, NAE, and IOM. 2010. Rising above the gathering storm, revisited: Rapidly approaching category 5. Washington, DC: The National Academies Press. https://doi.org/10.17226/12999.

NASEM (National Academies of Sciences, Engineering, and Medicine). 2016a. Gene drives on the horizon: Advancing science, navigating uncertainty, and aligning research with public values. Washington, DC: The National Academies Press. https://doi.org/10.17226/23405.

NASEM. 2016b. Genetically engineered crops: Experiences and prospects. Washington, DC: The National Academies Press. https://doi.org/10.17226/23395.

NASEM. 2017a. Preparing for future products of biotechnology. Washington, DC: The National Academies Press. https://doi.org/10.17226/24605.

NASEM. 2017b. Securing advanced manufacturing in the United States: The role of manufacturing USA: Proceedings of a workshop. Washington, DC: The National Academies Press. https://doi.org/10.17226/24875.

NASEM. 2017c. Software update as a mechanism for resilience and security: Proceedings of a workshop. Washington, DC: The National Academies Press. https://doi.org/10.17226/24833.

NASEM. 2017d. Undergraduate research experiences for STEM students: Successes, challenges, and opportunities. Washington, DC: The National Academies Press. https://doi.org/10.17226/24622.

NASEM. 2018a. Biodefense in the age of synthetic biology. Washington, DC: The National Academies Press. https://doi.org/10.17226/24890.

NASEM. 2018b. Graduate STEM education for the 21st century. Washington, DC: The National Academies Press. https://doi.org/10.17226/25038.

NASEM. 2018c. Indicators for monitoring undergraduate STEM education. Washington, DC: The National Academies Press. https://doi.org/10.17226/24943.

NASEM. 2019. Minority serving institutions: America’s underutilized resource for strengthening the STEM workforce. Washington, DC: The National Academies Press. https://doi.org/10.17226/25257.

NIH (National Institutes of Health). 2018. Foreign influences on research integrity. 117th Meeting of the Advisory Committee to the Director, December 13. https://acd.od.nih.gov/documents/presentations/12132018ForeignInfluences.pdf (accessed August 12, 2019).

Noyce, R. S., S. Lederman, and D. H. Evans. 2018. Construction of an infectious horsepox virus vaccine from chemically synthesized DNA fragment. PLoS One 13:e0188453.

NRC (National Research Council). 1996. Understanding risk: Informing decisions in a democratic society. Washington, DC: National Academy Press. https://doi.org/10.17226/5138.

NRC. 2001. Opportunities in biotechnology for future Army applications. Washington, DC: National Academy Press. https://doi.org/10.17226/10142.

NRC. 2004. Capturing the full power of biomaterials for military medicine: Report of a workshop. Washington, DC: The National Academies Press. https://doi.org/10.17226/11063.

NRC. 2012. Research frontiers in bioinspired energy: Molecular-level learning from natural systems: Report of a workshop. Washington, DC: The National Academies Press. https://doi.org/10.17226/13258.

NSB and NSF (National Science Board and National Science Foundation). 2018. Science and engineering indicators 2018. NSB-2018-1. Alexandria, VA: National Science Foundation. https://www.nsf.gov/statistics/indicators (accessed October 16, 2019).

Ochs, C., J. Geller, Y. Perl, and M. A. Musen. 2016. A unified software framework for deriving, visualizing, and exploring abstraction networks for ontologies. Journal of Biomedical Informatics 62:90–105. doi: 10.1016/j.jbi.2016.06.008.

Oliveira, A. L. 2019. Biotechnology, big data and artificial intelligence. Biotechnology Journal 14(8):1800613.

Peccoud, J., J. E. Gallegos, R. Murch, W. G. Buchholz, and S. Raman. 2018. Cyberbiosecurity: From naive trust to risk awareness. Trends in Biotechnology 36:4–7. doi: 10.1016/j. tibtech.2017.10.012.

Powell, P. 2008. Heparin’s deadly side effects. Time, November 13. https://web.archive.org/web/20121216232624/ http://www.time.com/time/magazine/article/0,9171,1858870-1,00.html (accessed July 5, 2019).

Reif, R. 2019. Letter to the MIT community: Immigration is a kind of oxygen. MIT News Office, June 25. http://news.mit.edu/2019/letter-community-immigration-is-oxygen-0625 (accessed July 3, 2019).

Richardson, L. C., N. D. Connell, S. M. Lewis, E. Pauwels, and R. S. Murch. 2019. Cyberbiosecurity: A call for cooperation in a new threat landscape. Frontiers in Bioengineering and Biotechnology 7:99. doi: 10.3389/fbioe.2019.00099.

Robinson, G. E., C. M. Grozinger, and C. W. Whitfield. 2005. Sociogenomics: Social life in molecular terms. Nature Reviews Genetics 6(4):257–270.

Ronquillo, J. G., J. E. Winterholler, K. Cwikla, R. Szymanski, and C. Levy. 2018. Health IT, hacking, and cybersecurity: National trends in data breaches of protected health information. JAMIA Open 1(1):15–19.

Saey, T. H. 2018. Genealogy databases could reveal the identity of most Americans. Science News, October 12. https://www.sciencenews.org/article/genealogy-databases-could-reveal-identity-most-americans (accessed October 16, 2019).

Shabani, M., and P. Borry. 2018. Rules for processing genetic data for research purposes in view of the new EU General Data Protection Regulation. European Journal of Human Genetics 26(2):149–156. doi: 10.1038/s41431-017-0045-7.

Shields, M. 2017. ChemChina clinches landmark $43 billion takeover of Syngenta. Reuters, May 5. https://www.reuters.com/article/us-syngenta-ag-m-a-chemchina/chemchina-clinches-landmark-43-billion-takeover-of-syngenta-idUSKBN1810CU (accessed September 5, 2019).

Smith, S., and J. Marchesini. 2007. The craft of the system security. Boston, MA: Person Education.

Stark, Z., L. Dolman, T. A. Manolio, B. Ozenberger, S. L. Hill, M. J. Caulfied, Y. Levy, D. Glazer, J. Wilson, M. Lawler, T. Boughtwood, J. Braithwaite, P. Goodhand, E. Birney, and K. N. North. 2019. Integrating genomics into healthcare: A global responsibility. American Journal of Human Genetics 104(1):13–20.

Taylor, D. O. Forthcoming. Patent eligibility and investment. SMU Dedman School of Law Legal Studies Research Paper No. 414. Cardoza Law Review. http://dx.doi.org/10.2139/ssrn.3340937 (accessed November 1, 2019).

Toga, A. W., and I. D. Dinov. 2015. Sharing big biomedical data. Journal of Big Data 2:7. doi: 10.1186/s40537-015-0016-1.

Tollefson, J. 2019. Chinese Americans uneasy as tensions disturb research. Nature 570:13.

Tucker, T. 2019. U.S. Army making synthetic biology a priority. Government Executive, July 1. https://www.govexec.com/defense/2019/07/us-army-making-synthetic-biologypriority/158145 (accessed July 3, 2019).

Tumpey, T. M., C. F. Basler, P. V. Aguilar, H. Zeng, A. Solórzano, D. E. Swayne, N. J. Cox, J. M. Katz, J. K. Taubenberger, P. Palese, and A. García-Sastre. 2005. Characterization of the reconstructed 1918 Spanish influenza pandemic virus. Science 310(5745):77–80.

U.S. Chamber of Commerce. 2017. Made in China: Global ambitions built on local protections. https://www.uschamber.com/sites/default/files/final_made_in_china_2025_report_full.pdf (accessed October 19, 2019).

U.S. Department of Commerce. 2018. Review of controls for certain emerging technologies. Federal Register 83(Nov. 19):58201–58202.

USGCRP (U.S. Global Change Research Program). 2018. Impacts, risks, and adaptation in the United States: Fourth national climate assessment, Vol. II. Washington, DC: U.S. Global Change Research Program. doi: 10.7930/NCA4.2018.

USPTO (U.S. Patent and Trademark Office). 2014. 2014 Interim guidance on patent subject matter eligibility. Federal Register 79:74618–74633. https://www.govinfo.gov/content/pkg/FR2014-12-16/pdf/2014-29414.pdf (accessed September 6, 2019).

USPTO. 2015. July 2015 update: Subject matter eligibility. Alexandria, VA: USPTO. https://www.uspto.gov/sites/default/files/documents/ieg-july-2015-update.pdf (accessed October 19, 2019).

USPTO. 2017. Patent eligible subject matter: Report on views and recommendations from the public. July. https://www.uspto.gov/sites/default/files/documents/101-Report_FINAL.pdf (accessed September 6, 2019).

USPTO. 2019a. 2019 Revised patent subject matter eligibility guidance. Federal Register 84(4):50–57. https://www.govinfo.gov/content/pkg/FR-2019-01-07/pdf/2018-28282.pdf (accessed September 6, 2019).

USPTO. 2019b. October 2019 update: Subject matter eligibility (issued October 17, 2019). https://www.uspto.gov/sites/default/files/documents/peg_oct_2019_update.pdf (accessed October 17, 2019).

USTR (U.S. Trade Representative). 2018a. Findings of the investigation into China’s acts, policies, and practices related to technology transfer, intellectual property, and innovation under Section 301 of the Trade Act of 1974. Executive Office of the President. https://ustr.gov/sites/default/files/Section%20301%20FINAL.PDF (accessed March 12, 2020).

USTR. 2018b. Update concerning China’s acts, policies and practices related to technology transfer, intellectual property, and innovation, November 20. Executive Office of the President. https://ustr.gov/sites/default/files/enforcement/301Investigations/301%20Report%20Update.pdf (accessed March 12, 2020).

USTR. 2019. 2018 Report to Congress on China’s WTO compliance. https://ustr.gov/sites/default/files/2018-USTR-Report-to-Congress-on-China%27s-WTO-Compliance.pdf (accessed September 5, 2019).

Van Noorden, R. 2013. White House announces new U.S. open-access policy. Nature Newsblog, February 22. http://blogs.nature.com/news/2013/02/us-white-house-announcesopen-access-policy.html (accessed September 5, 2019).

Vest, C. M. 2013. Testimony before the U.S. House of Representatives Committee on Science, Space and Technology, Subcommittee on Oversight, “Espionage Threats at Federal Laboratories: Balancing Scientific Cooperation while Protecting Critical Information,” May 16. https://docs.house.gov/meetings/SY/SY21/20130516/100836/HHRG-113-SY21-Wstate-VestC-20130516.pdf (accessed March 12, 2020).

Vlassov, V. 2007. Russian clinical research is threatened by ban on export of samples. British Medical Journal 334(7606):1237. doi: 10.1136/bmj.39241.621863.DB.

Wee, S. 2019. China uses DNA to track its people, with the help of American expertise. The New York Times, February 21. https://www.nytimes.com/2019/02/21/business/chinaxinjiang-uighur-dna-thermo-fisher.html (accessed October 16, 2019).

White House Office of the Press Secretary. 2013. Presidential policy directive—critical infrastructure security and resilience. https://obamawhitehouse.archives.gov/the-press-office/2013/02/12/presidential-policy-directive-critical-infrastructure-security-and-resil (accessed October 16, 2019).

Williams, P. A., and A. J. Woodward. 2015. Cybersecurity vulnerabilities in medical devices: A complex environment and multifaceted problem. Medical Devices (Auckland, N.Z.) 8:305–316. doi: 10.2147/MDER.S50048.

Zaveri, M. 2019. Wary of Chinese espionage, Houston cancer center chose to fire 3 scientists. The New York Times, April 22. https://www.nytimes.com/2019/04/22/health/md-anderson-chinese-scientists.html (accessed June 18, 2019).

Zhang, S. 2016. DNA got a kid kicked out of school—and it’ll happen again. Wired, February 1. https://www.wired.com/2016/02/schools-kicked-boy-based-dna (accessed October 16, 2019).

Zheng, S. 2019a. China issues official warning to students hoping to go to U.S. South China Morning Post, June 3. https://www.scmp.com/news/china/diplomacy/article/3012884/ china-issues-official-warning-students-hoping-go-us (accessed July 3, 2019).

Zheng, S. 2019b. Chinese told to raise safety awareness in U.S. travel advisory. South China Morning Post, June 4. https://www.scmp.com/news/china/diplomacy/article/3013046/ chinese-travellers-told-raise-safety-awareness-us-travel (accessed July 3, 2019).