The context of data sharing in genomics research

Data sharing is the practice of making research outputs available to other researchers by adding or combining research participants’ data into larger repositories [1]. The nuances between genomic data sharing and transfer include: data sharing involves making genomic information, such as DNA sequences, available to a wider scientific community to promote collaboration and scientific discovery. It also refers to making data accessible for use by others beyond the research team, often through controlled or open-access repositories, collaborative platforms, or federated systems. Data sharing emphasizes enabling research, reproducibility, and collaboration while considering privacy, consent, and governance frameworks. In contrast, data transfer refers to the act of moving genomic data from one location to another, such as between servers, institutions, or individuals, with the goal of facilitating data for analysis or interpretation [2]. In addition, data transfer entails the point-to-point movement of data which could be the physical or digital movement of data from one entity or jurisdiction to another. Formal agreements and adherence to data protection laws may be required to enable this process, which is often subject to legal and regulatory scrutiny particularly in cross-border contexts. Such agreements typically define the parties involved, the specific systems, and the operational purposes (for example, data storage, processing, or analysis) without granting broader access rights. An increasing number of government departments, research communities, funding agencies and scholarly journals are developing initiatives and policies to promote data sharing and greater access to data, recognizing their enormous potential for scientific, social, and economic growth [3–6]. Open data policies from European countries [7,8] and the United states of America Wheeland [9] increasingly require custodians of human genomic data to make it as widely available as feasible [10]. Data sharing is regarded as essential for enabling and promoting genomic research in a way that will maximize the benefits to public health [11]. Additionally, several frameworks guiding the sharing of genomic data have been developed some of which are: the Genomic Data Sharing Policy [12], International Declaration on Human Genetic Data [13], International code of conduct for genomic and health related data sharing [14] and the Framework for Responsible Sharing of Genomic and Health Related Data [15]. The frameworks indicate that the rights of participants and their communities should be protected, they emphasize the importance of informed consent, de-identification of the shared data, privacy and confidentiality, sharing benefits from the study with the community where the participants were recruited from and ensuring everyone has access to the shared data, placing mechanisms in place to ensure that the shared data is not used to discriminate and stigmatize the individuals. They further indicate that transparency and trust are key components in the sharing of genomic data.

Genomic research is increasingly becoming common with researchers participating in different projects required to share details of their participant data. The sharing of genomic data promises to increase research efficiency, expedite translation of research results, and ensure the traceability and transparency of published studies and maximize the utility of results [16–19]. Research funders and sponsors demand that genomic sequences be deposited on public data repositories unless there are justifiable reasons why this should not be so [20]. The need for broad access to genomic data brings along a host of ethical concerns, including those related to privacy, confidentiality as well as fairness and equity [21]. The risk of re-identifiability remains a major ethical concern in genomic data sharing, even with the implementation of technical safeguards such as controlled-access repositories, encryption, pseudonymisation, and secure data transfer protocols [22–24]. Challenges of effective data sharing include: 1) absence of established standards for data users, 2) researchers from low and middle income countries (LMICs) often experience inequities in collaborative research, including not being appropriately credited for their contributions, 3) loss of intellectual property rights, 4) misuse of data [25,26] and 5) absence of benefit sharing frameworks [27,28]. These rooted disparities highlight the need to decolonize research ethics by promoting fair authorship practices, shared governance, and more equitable participation in global health research [29,30]. An analysis of genomic guidelines, policies, and procedures from LMICs by Ali et al 2021 [31] and de Vries [32] revealed significant weaknesses and gaps in the governance of genomic research and biobanking [31,33]. Several LMICs have enacted legislation through Data Protection Acts to protect personal data, however these laws often lack sector-specific provisions and detailed guidance for health research, presenting a significant challenge [34]. Most sub-Saharan countries lack ethical and legal frameworks to guide and regulate the sharing of genomic research data. The shared genomic data is inclusive of both genotypic and phenotypic data. Discussions should be held with researchers over whether data providers should review results before publication, collaborate on the analysis, approve the analysis plan in advance which are important areas for dialogue on data sharing [26].

Although data sharing enhances efficiency and holds great promise for advancing scientific knowledge and improving public health outcomes, it also presents a range of complex ethical, legal, and social issues (ELSI). These challenges include concerns around privacy, informed consent, data ownership, equitable access, and the potential for misuse or discrimination, all of which are especially significant in the context of genomic research [21,35–40]. One of the key challenges is determining how to protect the privacy of participants while enabling the sharing of data through global research networks [41]. Several concerns regarding privacy, discrimination, data misuse, inequity and stigmatization among many have led to policy responses from the National Human Genome Research Institute and additional policies from domestic and international countries that reaffirmed the recommendations for publicly sharing genomic data [17]. These policies restrict access and occasionally limit the utility of data to genomic data as a means of protecting research participants [17]. Only bona fide individuals with appropriate institutional credentials are granted access to permissioned data for ethically approved purposes under this policy. To promote ethical and equitable data sharing in genomic research, there is need for appropriate regulatory frameworks. However, the development of such frameworks should take into consideration the perspectives of the various stakeholders in the genomic research landscape to ensure that they are locally appropriate and context-specific.

Despite the valuable contributions made by earlier reviews, they remain limited in scope. Many of them focus on African or high- income settings and primarily examine policies, guidelines, and public attitudes toward genomic data sharing, with some restricted to pediatric contexts. For instance, a review by Ali et al. [31], mapped ethics-related guidelines and legal instruments governing genetic and genomic research across African countries, providing a policy-level overview rather than an analysis of data-sharing practices [31]. Rahimzadeh, Knoppers, and Bartlett [42] reviewed ethical, legal and social issues related to sharing whole-genome data, but their focus was restricted to pediatric genomics, thereby excluding broader stakeholder and institutional perspectives. More recently, Amoakoh-Coleman et al. [43] described ethical considerations for biobanking and genomic data use in Africa; however, their emphasis remained on biobank establishment, consent frameworks and regulatory gaps, without examining how data are shared in practice or how contextual factors influence data access, reuse or governance [43]. A review by Shabani Identified key concerns influencing willingness to share data, but did not examine how these attitudes interact with governance systems, but did not examine how these attitudes interact with governance systems, institutional practices or consent implementation in real settings, particularly in LMIC collaborations [44].

This review collated evidence on data sharing practices, context, facilitators and barriers in collaborative human genomic research in LMICs. Moreover, the review is justifiable because it was conducted to inform ethical data or guide collaborative human genomic research and it will enhance the body of evidence on ethical, legal and social implications of data sharing in collaborative genomic research setting criteria for data sharing. This review broadens the scope by examining genomic data sharing in Africa, Asia and other LMICs, alongside collaborations with the UK, US, Canada and Europe, allowing comparison of governance challenges in both Global North–South and South–South partnerships.

Protocol development

A protocol was developed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist recommended for systematic reviews [45–48]. This protocol was further developed and subsequently published in PLoS ONE [49].

Definition of LMICs

In this review, LMICs are defined according to the World Bank’s 2024 country income classification, which categorizes economies based on gross national income (GNI) per capita. Countries with a GNI per capita below US$13,846 were considered LMICs [50,51]. While a substantial proportion of the included studies were conducted in African countries, the review also incorporates research from Asia and other LMIC regions.

Review question

What is the context of sharing human genomic data in collaborative research in LMICs between 2003–2025?

The elements of the population, intervention, context, outcomes, setting and timeframe (PICOST) are presented in Table 1.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Elements of PICOST.

https://doi.org/10.1371/journal.pone.0354471.t001

Time justification

The review focused on research conducted from 2003 onward, marking the completion of the Human Genome Project [41,52]. This period was selected because the project highlighted key lessons such as the importance of collaboration, data sharing, and the ethical management of genomic information that have since shaped how genomic research is designed and conducted.

Eligibility and selection of papers

Data sources

The literature review involved a search of three electronic databases, namely PubMed, Google Scholar, and Web of Science in line with Cochrane Handbook recommendations to search at least two to three databases for comprehensive coverage [53–55].

Inclusion criteria

All articles that met the PICOST criteria (Table 1), as defined by the research question, were included and constituted the unit of analysis. In addition, articles were included if they focused on data sharing in genomics and reported original research; both published and grey literature were considered. Date restrictions were applied to the initial electronic search to include articles published between 2003 and 2025, with the search limited to publications and we used English terms for search. Studies were included if they involved research collaborations between institutions in LMICs and those in high-income countries (Global North–South partnerships) or among LMICs themselves (South–South collaborations).

Exclusion criteria

Papers were excluded if they focused on personal health records, clinical results, letters, opinion papers and articles that only focused on biobanking and not data sharing. We excluded papers that did not report relevant outcomes on genomic data sharing and those that did not stratify results for LMICs [49].

Data abstraction and coding

Data extraction was done using a pre-designed and piloted form in Microsoft office excel [56] and EPPI reviewer software by a pair of reviewers, acting independently to capture the following information from included articles: author, year of study, country/region, study design and relevant outcomes [57]. The lead reviewer (DES) with another member of the team conducted data extraction in duplicate for a proportion of papers to enhance quality of the process. To assess the consistency of data extraction, a subset of papers was independently reviewed by two team members. The EPPI-Reviewer report [58] using the coding reports indicated a 92% level of agreement. Any discrepancies were discussed and resolved through consensus, further enhancing the rigor and credibility of the synthesis. Results of the full text extraction were shared with the remaining review authors to validate them.

Coding and synthesis

A narrative synthesis was used to integrate and interpret findings from multiple included studies [59]. This approach was chosen due to the diversity in study designs, settings, and outcomes, which made quantitative synthesis inappropriate. We were guided by the framework proposed by Popay et al. [59] to ensure a systematic and transparent process. Data synthesis was conducted using a structured narrative approach, enabling the systematic identification and interpretation of key patterns across the included studies [60]. Two reviewers (DES and a research assistant) independently read and re-read each study to become thoroughly familiar with the content and to identify recurring themes and insights. The synthesis primarily relied on textual analysis to summarize and explain findings. As part of the preliminary synthesis, the reviewers gathered and arranged the findings from each study in a way that made it easier to see commonalities, differences, or trends across the studies. Key data such as study aims, methodology, population, and outcomes were extracted into a matrix developed in Microsoft Excel. This matrix also included relevant quotes and summary statements to preserve the context and depth of each study.

Handling of missing data

Variables that were desired but missing or not reported were denoted as not reported (‘NR’) and clarification was sought by contacting the study authors. Papers with essential data that remained unavailable due to author nonresponse were excluded, as the missing information was necessary to determine eligibility and ensure methodological rigour. Excluding these studies ensured that the final evidence base was complete, verifiable and suitable for a reliable synthesis.

Quality assessment

A publication assessment of all the included studies was done following the Joanna Briggs Institute (JBI) guidance standardized critical appraisal tool for assessing methodological quality [61]. Based on the total scores, studies were classified as good (8–10), fair (5–7), or poor (0–4), following published guidance and prior reviews using similar thresholds, ensuring consistency and transparency in our assessment of methodological rigor. This scoring involved evaluating how rigorously the research was conducted, including data collection and analysis methods. It was done to ensure the included studies were credible, trustworthy, and reliable. To enhance transparency in the synthesis, the appraisal scores were used to guide the inclusion and interpretation of evidence. Studies rated as poor were excluded from the review to ensure that only credible and methodologically sound evidence informed the findings.

PRISMA flow chart of data sharing in human genomics collaborative research

A summary of the findings indicates that 2,091 records were identified through database searches: 1,127 from Google Scholar, 340 from Web of Science, 594 from PubMed, and 30 from targeted sources. After removing 650 duplicates, 1,441 records remained for title and abstract screening. Following full-text review of 98 articles, 88 were excluded for not meeting the inclusion criteria. Ultimately, 11 studies were included in the narrative synthesis (see Fig 1).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Showing the PRISMA flow chat.

https://doi.org/10.1371/journal.pone.0354471.g001

Description of articles

A total of 11 studies were included in the systematic review of which nine were qualitative studies, one quantitative study and a mixed method study (Table 1). The qualitative articles included [32,62–69] and a mixed method article [70] and the quantitative article [25]. The studies focused on collaboration and data sharing in LMICs as well as on the sharing of human genomic data.

The characteristics of the included studies have been summarized in Table 2.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Characteristics of included studies.

https://doi.org/10.1371/journal.pone.0354471.t002

A summary of the qualitative articles which were included in the review are summarized showing the themes and quotes (Table 3).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Summary of the qualitative studies.

https://doi.org/10.1371/journal.pone.0354471.t003

Publication assessment of the studies

We assessed the methodological quality of all 11 included studies in the systematic review. The assessment comprised qualitative, quantitative, and mixed-methods research, and the JBI appraisal checklist was used (see appendix 4, Table I and 4.1, Table 2) [61,68]. (Appendix 4, Table 1 for qualitative studies; Appendix 4.2, Table 2 for quantitative studies). The quality assessment yielded overall ratings of good indicating that the included studies demonstrated sound methodological rigor. None of them were rated as poor and because of this, we did not exclude any study from our analysis. Instead, we brought all the findings together, while keeping in mind the differences in how rigorous each study was when interpreting the evidence.

Ethics approval and consent to participate

All empirical studies included in this review reported receiving the required ethical approval from their respective institutional review boards or ethics committees.

Additionally, this systematic review was approved by the Makerere University Higher degrees committee School of Biomedical Sciences Research ethics committee (SBSHD-REC 2022−273) and the Uganda National Council for Science and Technology (SS1730ES).

Narrative synthesis

The narrative synthesis produced seven overarching domains, developed through iterative discussion and reflection to capture both frequent and salient findings. These domains were: 1) context of data sharing, 2) genomic data sharing practices, 3) ethical concerns, 4) social dimensions, 5) legal concerns, 6) facilitators, and 7) barriers.

Strengths and limitations

This review offers valuable insights into the context, facilitators and barriers to the sharing of human genomic data sharing by leveraging qualitative evidence.

While it effectively highlights key factors such as data de-identification, responsible sharing practices, and the presence of data protection frameworks. A quality assessment was conducted of the studies which were included in the review which promotes the reliability and generalizability of the findings.

This review adds to the growing body of knowledge on the ELSI in genomic data sharing, highlighting the need for further inquiry to help protect the rights of both researchers in LMICs and research participants.

The findings of this review will guide regulators and policy makers in determining the best how to protect the personal interests of research participants while enabling the sharing of data through Global research networks.

Our reporting is aligned to the PRISMA statement [83,84] and the full report will be shared with relevant stakeholders including universities, civil society, funders, and departments of genomic and genetic research to ensure an adequate reach especially in LMICs. The high level of agreement (92%) demonstrates strong inter-rater reliability and suggests that the data extraction process was applied consistently.

The review conducted a comprehensive search of relevant papers up to the year 2025. We synthesized data from a wide range of databases which provides data from a relevant scope [43]. This ensures that the findings are based on the most recent and relevant literature available. This context is crucial for understanding the regulatory environment that influences data sharing practices.

A limitation of our review is its primary focus on qualitative studies, which meant that only a small number of quantitative articles were included which may affect generalizability. However, we carefully examined their methodologies to ensure they demonstrated sufficient rigour before incorporating their findings. In addition, the review concentrated on studies conducted in LMICs, with a particular emphasis on African contexts, which may limit the generalizability of the findings to other regions. The use of a narrative synthesis approach also presents limitations, as it relies heavily on reviewer judgement when interpreting and organizing themes, introducing the potential for subjective interpretation and bias. Finally, we did not conduct a meta-analysis due to the heterogeneity of study designs and outcomes.

Overall, our comprehensive approach in reviewing only a few articles as well as inclusion of recent literature provides an indepth understanding of the context and ethical, legal, and practical considerations in the sharing of human genomic data. This approach enabled us to engage critically with each study, identify indirect but important themes, and generate a richer, more context-sensitive synthesis than would have been possible with a broader but less detailed review.

Reporting and dissemination

Our reporting follows the PRISMA statement guidelines [83,84], which emphasize transparency and openness in research and are aligned with the principles of the Open Science Framework [85]. A policy engagement meeting is underway, and the full report will be disseminated to key stakeholders, including universities, civil society organizations, funders and departments responsible for genomic research. This will also involve community engagement to ensure broader public understanding and participation. In addition, preliminary results from the review have been disseminated at both local and international conferences. Together, this dissemination strategy is intended to ensure that the findings reach relevant decision makers and contribute to ongoing policy development and practice in LMIC settings.

Appendix

An assessment of the methodological quality of the included studies is presented in Table 4.

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Quality assessment of the included studies (n = 10).

https://doi.org/10.1371/journal.pone.0354471.t004