Overview

The rapid proliferation of life science data presents tremendous opportunities to address long-standing and emerging challenges in health, agriculture, the environment, and other sectors. Data from scientific research represent the outcomes of significant investment on the part of individuals and communities, institutions, industries, and governments. Some data generated and collected are sensitive, often warranting special protections. Life science data can include, for example, personal and genetic data, proprietary data, and/or data that may be otherwise harmful if accidentally or intentionally misused. Different types of life science data (e.g., genetic vs. nongenetic, human vs. animal) may necessitate different considerations to appropriately balance risks and benefits of data sharing and protection. To facilitate scientific progress and collaboration while also protecting personal privacy, the integrity of research investments, and national security, it is important for life science data to be collected, digitized, secured, and shared responsibly.

To exchange experiences and collaboratively discuss best practices for protecting life science data and promoting responsible data sharing in Central Asia, the U.S. National Academies of Sciences, Engineering, and Medicine convened a workshop series titled “Engaging Scientists in Central Asia on Life Science Data Governance Principles,” held virtually across six 2-hour sessions in May and June 2023. The workshop series brought together early-career and established scientists, along with experts in data science, cybersecurity, and law, from Kazakhstan, Kyrgyzstan, Tajikistan, the United States, Uzbekistan, and elsewhere to explore existing policies and practices, gaps and limitations, fundamental concepts relevant to data governance, and opportunities for improvement in data sharing practices and governance structures. Throughout the workshop series, participants considered how concepts and approaches relevant to data governance fit into the overall research context to help minimize the risks and maximize the benefits of life science research endeavors in Central Asia and elsewhere. In addition, the National Academies commissioned seven papers by authors from Kazakhstan, Kyrgyzstan, and Tajikistan on a variety of issues related to data governance. These papers are provided in Appendix E of this report.

Data governance has been defined as “the principles, procedures, frameworks, and policies that ensure acceptable and responsible processing of data at each stage of the data life cycle… [to] maintain data integrity, quality, availability, accessibility, usability, and security” (Eke et al., 2022, p. 600-601). Data governance is important not only because researchers have a responsibility to protect personal privacy and the integrity of scientific investments, but also because good data governance makes science more impactful. Each stage of the data life cycle—including data collection, storage, processing, curation, sharing, application and use, and deletion—has unique considerations that impact data governance. For instance, during data collection, researchers consider such issues as whether informed consent for sharing and using the data has been obtained, whether biases exist for sampling, and whether there are legal and/or regulatory differences that affect sharing the data between jurisdictions. During data storage, considerations include how to safely and securely collect and handle data, how long data should be preserved, who owns the rights to the data, who pays for data storage, and how to minimize risks of data leaks as technology changes over time. In the application and use of data, researchers consider how to communicate incidental findings that pertain to the health of study participants, how to

ensure that the minimal amount of data is used to reduce risk to participants, how to ensure data are not misused, and how to ensure data are used responsibly for civil and military applications or for economic gain (Eke et al., 2022). “Good governance frameworks really achieve two aims: They maximize the scientific utility of that data, as well as the social value of the research that’s derived from that data,” said committee member Vasiliki Rahimzadeh, Baylor College of Medicine (United States). She noted that responsible governance frameworks help governments and institutions use the outcomes of scientific research to make informed decisions for the benefit of all members of society.

By providing a forum for open exchange of information, experiences, and views on data sharing and protection, the workshop series helped elucidate several challenges and opportunities facing researchers who seek to maximize the value of their research investments for scientific progress while upholding values such as respect, fairness, safety, and security. “Now more than ever, life scientists are collecting and analyzing vast amounts of data from various sources,” said Lusine Poghosyan, Columbia University (United States), chair of the workshop’s planning committee. “Digitalization of these data presents new opportunities to address significant challenges across various sectors and accelerate research through collaboration within and across borders. It’s important to ensure that data are shared responsibly, and that appropriate safeguards are in place to protect the privacy of individuals and the investments made by institutions, industries, governments, and others.”

In sessions highlighting particular scientific topics being pursued in Central Asia, participants explored both the benefits of scientific advances and the risks and vulnerabilities involved in collecting and sharing biological data and knowledge. In the first workshop, participants presented a sampling of life science research efforts in Central Asia. Scientists from Kazakhstan, Kyrgyzstan, and Uzbekistan, working in veterinary medicine, infectious disease, antimicrobial resistance, genomics, botany, and other areas, shared their successes and challenges in conducting life science research, digitizing data, and contributing to international databases. They pointed to some key challenges facing the region’s research efforts, with common limitations including a dearth of comparison data from the region, difficulty accessing biobanks and other technological resources, and limited expertise in some areas such as data science.

While there are currently no international data governance laws, several organizations offer resources and guidelines for responsible data sharing across borders. These include the United Nations Educational, Scientific and Cultural Organization; the World Health Organization; and the Global Alliance for Genomics and Health. At the second workshop, several participants suggested that successful international collaborations hinge on developing and adopting data interoperability standards and harmonizing national data sharing and security laws. Participants examined existing international and national policies, practices, and norms related to access and transfer of biological and life science data, focusing on data sharing and protection. Data governance experts from Germany, India, Taiwan, Uganda, and the United States described their data sharing laws and norms as potential models for researchers in other countries to learn from when developing, evaluating, and improving upon their own frameworks. Some participants also attested that their current national data governance standards are in flux. During these discussions, speakers also described challenges in data management and international data sharing, such as the cost of long-term data storage and the lack of advanced infrastructure and subject matter expertise in some regions.

The goal of the third workshop was to provide a balanced view of biological data sharing. Globally, there is a shift toward a culture of open science, which is widely viewed as the key to advancing life science research. However, in practice, hurdles—such as real and perceived risks and harms for participants—prevent biological data from being shared freely and openly. In the third workshop, the benefits of sharing biological data were considered together with the risks and vulnerabilities of doing so, in order to explore how data can be shared more feasibly, equitably, and securely. As an example, several speakers noted that tribal and Indigenous communities worldwide can be harmed by unethical research practices; in response, Indigenous communities are exploring methods of protecting and stewarding the data and knowledge their communities hold, in order to have greater input into decisions on how data from their community members are collected, stored, and shared, and to have a greater share of the benefits derived from research using those data and knowledge, especially when the benefits directly affect their communities. Each of the five Central Asian countries has communities with unique traditional knowledge, which not only holds considerable personal and spiritual meaning but also may be important for the survival of these communities. Several speakers suggested that stronger legal protections may be needed to preserve traditional knowledge and noted that access to resources and legal expertise could help address inequities in the ability for communities to navigate intellectual property and patent law. Participants discussed different risk, use, and sharing considerations that apply to a variety of data types. Different perspectives on the definitions of and roles for ethics, equity, and equality in science were also shared.

Participants also explored life science data governance principles in practice. In the fourth workshop, speakers shared experiences with the existing national practices, policies, and norms around protecting and sharing life science data in Central Asia; described unique features, common challenges, and common goals; and explored opportunities to improve data governance policies and implementation. Existing practices address electronic data governance, conflicts of interest, personal data, biometric registration, genetic resources, and centralized digitization efforts. These policies and norms are often applied in a project-centric way, and researchers are typically required to secure their data and publish or deposit them into international repositories. Emphasizing the importance of standardized protocols and appropriate security measures to support data sharing, panelists from Kyrgyzstan, Tajikistan, and Uzbekistan identified common challenges with data sharing in their countries, including limitations in available data and computing power; a lack of expertise in cybersecurity, artificial intelligence, and digitization; and a need for improved digital infrastructure.

The workshops also surfaced opportunities and challenges in terms of the ability to translate data governance practices in different institutional contexts. The fifth workshop focused on managing cybersecurity risks with the goal of maintaining the confidentiality, integrity, and availability of life science data. Speakers commented that a commitment to open science includes a responsibility to share data in a secure and meaningful way to advance science, but the methods employed to achieve this vary depending on an institution’s context, goals, and resources. Cyberattacks can disrupt research and compromise sensitive data, and it is therefore important to balance open data sharing with risk assessment, especially when international collaborations or dual-use research are involved. Speakers discussed several approaches to advance cybersecurity in scientific organizations, including basic cyber hygiene, resources, and practices, along with more intensive cyber resilience strategies. Panelists also emphasized the importance of vigilance in the face of

social engineering methods, such as phishing, that can manipulate staff into divulging sensitive information, such as passwords and personal information.

In the sixth and final workshop, organizers summarized what was learned from the previous five workshops and facilitated a collaborative discussion to explore areas on which to build. Participants highlighted concrete suggestions that decision-makers in Central Asian countries can implement to improve their data governance frameworks, infrastructure, and expertise, thereby allowing researchers to participate in international collaborations more fully. One recurring theme throughout the workshop series was the idea that researchers have a duty to share data responsibly in an effort to advance science and that this means recognizing cybersecurity threats and implementing appropriate access controls and data protections. Working with, protecting, and sharing data require time and resources, however. As Yann Joly (committee member), McGill University (Canada), stated, “the devil’s in the detail[s].” Joly reiterated that creating an agreed-upon international data governance framework requires work: integrating and harmonizing the disparate national laws; creating incentives to motivate scientists to follow responsible data sharing practices; and designing clear policies that protect subjects’ and researchers’ data, intellectual property, and national security.

Looking forward, workshop series planning committee members highlighted four ways participants and other experts can implement and advance the life science data governance best practices that were described during the workshop series. First, participants can create and support regional or national scientific networks consisting of data repositories, scientific societies, journals, and conferences, some of which already exist or are being built. Second, funding sources can be sought for in-person meetings and sustained engagement, initiatives, and collaborations. Third, Central Asian scientists can be encouraged to publish in journals and use repositories that are available internationally, with help from international researchers who can offer scientific communication training to overcome language or cultural barriers. Finally, engaging diverse groups of scientists, students, decision-makers, ministry officials, legal experts, and other relevant groups in creating a more supportive environment for scientific collaboration can help to expand the reach and benefits of research. Participants closed the final workshop session by discussing follow-on activities being planned to disseminate the information and views shared during the workshop series along with future opportunities to continue the conversation with a broader set of participants.