Abstract

Background:

Integrating biobanking and genomic research into health systems in low- and middle-income countries (LMICs) provides considerable opportunity to advance precision medicine and promote global health equity. However, persistent structural disconnect exists between the physical infrastructure for biospecimen collection and digital frameworks required to generate, manage, and share associated data. This biospecimen-data divide excludes LMIC populations from the benefits of genomic research and increases community vulnerability to extractive data policies and dependency on high-income-country (HIC) partners.

Objective:

This conceptual analysis aims to: (1) deconstruct the biospecimen-data divide by describing its core elements and interrelations; (2) introduce the DIGS (Digital Infrastructure, Interoperability, Governance, and Skills) Model as a framework for reconceptualizing biobanking as integrated digital enterprise; (3) examine how decisions regarding digital infrastructure may reproduce or reduce health inequities in LMIC settings.

Methods:

A critical scoping review was conducted using the Arksey and O’Malley framework and reported in line with PRISMA-ScR. Five databases—PubMed, Scopus, Web of Science, Embase, and Global Index Medicus—were systematically searched, yielding 2,390 records. After deduplication (n = 880), title and abstract screening (n = 1510), and full-text review (n = 296), 154 studies were included in the final synthesis. These included peer-reviewed articles, policy documents, and grey literature on biobanking, digital health infrastructure, and data governance in LMIC contexts.

Results:

Four interdependent structural gaps underpin the biospecimen-data divide: (1) infrastructure gap, reflecting constraints on data generation and flow; (2) interoperability gap, arising from incompatible systems that create data silos; (3) human-capital gap, marked by limited availability of personnel with combined laboratory and data stewardship expertise; (4) governance gap, defined by weak ethical and regulatory structures. Together, these gaps reveal deeper asymmetries between HICs and LMICs. The DIGS Model reframes biobanking as a cyclical, equitable process requiring coordinated investment across all four dimensions.

Discussion:

The DIGS Model challenges the prevailing infrastructure-first paradigm, which prioritizes physical assets over digital capabilities. It advances partnership models that center LMIC leadership in data governance, redefining sustainability as the cultivation of local capacity to generate, interpret, and control data. The framework offers researchers, funders, and policymakers a shared diagnostic tool for bridging the biospecimen-data divide without reinforcing existing dependencies.

1 Introduction1.1 The genomic revolution and its potential for LMICs

Recent advances in genomic science and technology have helped us understand the causes of many diseases, inspired new approaches in genomic medicine, and given us tools for precision public health at the population level. Infectious, non-communicable, and neglected tropical diseases are especially common in Low- and Middle-Income Countries (LMICs), where genomics could help reduce health inequalities and suffering. Although the cost of sequencing a genome has dropped from $95 million in 2001 to less than $1,000 in 2022, resource-poor settings still face major challenges with infrastructure and capacity. Those challenges are the main focus of this study (1–3).

Initiatives like the Human Heredity and Health in Africa (H3Africa) Consortium have helped promote genomic research for African populations. Current efforts concentrate on building capacity and guaranteeing fair data sharing (4, 5).

Pathogen genomics has shown strong potential for field surveillance and outbreak management, though rapid technical responses have not consistently produced lasting digital infrastructure. Gaps in data generation and governance show that technical progress alone, without investment in digital systems, governance, and skills, leads to fragile and unequal results (6). To completely realize the benefits of genomics for inclusive health and meet the United Nations Sustainable Development Goal 3, Good Health and Well-being, we need not only laboratory capacity but also digital systems that link biospecimens to data (7, 8).

1.2 Biobanks as critical infrastructure for digital health in LMICs

Biobanks are essential to the genomic and digital health sector. They systematically collect, process, store, and share biospecimens along with related clinical and molecular data. Rather than being static archives, biobanks act as active hubs that connect local communities, healthcare providers, and researchers (9). In LMICs, where health systems encounter overlapping challenges from infectious diseases, chronic conditions, and new threats, biobanks are key to using digital technologies for fairer health outcomes, as long as the technological side of this infrastructure is well developed (10).

Digital tools have become more common in biobanking, with systems such as biobank information management systems (BIMS) and laboratory information management systems (LIMS) helping automate biospecimen tracking, reduce errors, and improve data quality (11, 12). Cloud systems and virtual biobanks could enable institutions to search distributed repositories without moving physical samples, but in LMICs, these solutions are still new and rely on stable internet connections (13, 14). Importantly, digital progress is more than just using new devices. It means creating workflows so that each biospecimen is treated as a data-rich object from collection to analysis and reuse, following the biospecimen-as-data-carrier idea (15). Building local control over these digital systems is important to make sure the benefits from research reach the communities where the samples come from (10). Even with new technology, biobanks in LMICs still face big challenges that highlight both their importance and their vulnerability.

Infrastructure problems, like unreliable electricity and internet, slow down digitization in many LMICs. A survey by Mendy and colleagues found that about 55% of biobanking facilities in 23 LMICs had uninterrupted power (16). But this average hides big differences: facilities in upper-middle-income countries and large cities often have better reliability, while those in rural or low-income areas sometimes have less than 30% reliable power (15, 16). Various regions have seen small improvements, especially where solar backup systems are used, but there is still not much updated data across all LMICs (17). Limited budgets and not enough skilled staff make things harder, often forcing reliance on outside funding and cross-border partnerships. Gaps in ethics and regulations, especially concerning data privacy, can threaten participant trust and data security, especially within diverse cultural settings in which informed consent may be difficult (18).

Recent projects show that biobanks are playing a bigger part in promoting health equity. In Latin America, expanding biobanking has focused on involving communities and meeting local health needs, while also tackling funding and regulatory challenges to improve research on regional diseases (19). The Pan-African Biobanking Network’s 2024–2025 survey found progress in building repositories across Africa, even though there are still issues with standardization and visibility. These are being addressed through fair partnership models (20). The FIND Integrated Biobank Network shows how networked governance can boost efficiency and help LMICs prepare for pandemics (21). Together, these efforts show that the main barriers are not simply technical, but are deeper structural issues: the gap between biospecimens and data, which need coordinated investment in infrastructure, governance, and skills.

1.3 The central problem: the biospecimen-data divide

There is a clear imbalance in biobanking investments in LMICs. Most resources and genomic data examination capabilities are found in wealthier countries and institutions (22). This asymmetry has been characterized in the literature as a pattern of extractive genomic research: Researchers describe this as extractive genomic research, where biospecimens and data are taken from LMIC populations and sent to high-income countries (HICs), but local communities do not receive equal benefits, analytical skills, or control (23–25). While a lot is spent on physical infrastructure like freezers, lab equipment, and storage, digital systems regularly get much less attention (26). This focus misses important needs, such as systems for managing clinical data, platforms that connect biospecimens to health outcomes, ethical data-sharing structures, and trained staff to run these digital tools (27). As a result, biobanks in LMICs often focus on collecting and storing samples but have trouble using digital tools effectively. This leads to less useful research and ongoing health inequities (15, 28). This situation is known as the “biospecimen-data divide,” which describes the gap between collecting and storing samples and being able to use the related data for scientific progress (29).

In many LMICs, the biospecimen-data divide worsens due to weak infrastructure, which makes it hard to use key technologies such as BIMS and LIMS. Old IT systems from different projects make it difficult to connect data across repositories and link with health records or omics datasets (28). If biospecimens are not linked to clinical, demographic, or genomic data, they have little value for developing treatments or public health programs, turning them into unused collections (30). Sending biospecimens from LMICs to high-income countries for analysis, without fair sharing of data or benefits, increases global inequalities. For example, in Armenia, limited bioinformatics skills and regulatory issues prevent local data use and participation in international projects (31). Some new projects, like the DxConnect Virtual Biobank and drone-based sample transport in Rwanda (32, 33) show ways to close this gap with digital tools and better logistics. However, ongoing problems with funding (such as Armenia’s R&D spending at only 0.32% of GDP) (31) and lack of regulatory coordination still slow progress. In One Health approaches, this imbalance makes LMICs more vulnerable during disease outbreaks, as poor data sharing and inconsistent information weaken surveillance and response, even though these countries often face higher disease burdens (28). To close the biospecimen-data divide, LMICs need targeted investments in digital infrastructure, training for local experts, and ethical guidelines that match international standards like ISO 20387 and GDPR (34).

1.4 Heterogeneity within the LMIC category: LIC, LMIC, and UMIC distinctions

The category “low- and middle-income countries” encompasses 129 economies spanning an extraordinary range of income levels, state capacities, health system maturities, and digital infrastructure landscapes. The World Bank’s current income classifications distinguish between low-income countries (LICs) with Gross National Income (GNI) per capita ≤ USD 1,135, LMICs: GNI per capita USD 1,136–4,495, upper-middle-income countries (UMICs): GNI per capita USD 4,496–13,935, and HICs: GNI per capita >USD 13,935 (35). While this manuscript uses “LMICs” as a shorthand for the full range of non-HICs settings, the DIGS Model’s relevance and priority dimensions differ meaningfully across this spectrum.

1.5 Aims

This conceptual analysis argues that to bridge the gap between biospecimens and data in biobanking within LMICs, we need to shift our view of biobanks. Instead of seeing them mainly as storage facilities, we should see them as digital enterprises that connect biospecimens with advanced data systems (12, 22). We suggest that local researchers, policymakers, and funders should work together to reorganize biobanking strategies and support fair knowledge creation (23). To support this idea, the analysis has three parts. First, it breaks down the biospecimen-data gap by looking at key issues like data differences, errors before analysis, and problems with systems working together. Second, it introduces the DIGS Model (Digital Infrastructure, Interoperability, Governance, and Skills), which brings these parts together inside a framework that can work in settings with fewer resources. This model is based on data governance in digital health and points out the need for guidelines that protect privacy and security. Third, the analysis looks at how this new approach could affect research methods by fostering the use of AI to make better use of biospecimens. Finally, it discusses how virtual biobanks can help make global health research more inclusive and reduce gaps in biomedical progress (36, 37).

2 Methods2.1 Methodological framework

We conducted a critical scoping review using the five-stage framework by Arksey and O’Malley (38) updated by Levac et al. (39), and followed the PRISMA-ScR checklist (40). We chose a scoping review because our aim was to map concepts and build a framework. The review focused on three main research questions: (1) What are the main barriers to adding digital infrastructure to biobanking in LMICs, and how are these barriers connected? (2) What frameworks, governance models, and technical standards have been suggested or used to address these challenges? (3) How can a single framework bring together infrastructure, interoperability, governance, and skills to support fair biobanking in settings with limited resources?

2.2 Search strategy, database selection, screening, and eligibility criteria

We systematically searched five electronic databases: PubMed/MEDLINE, Scopus, Web of Science, Embase, and Global Index Medicus, covering the period from January 2010 to January 2026. We also searched targeted grey literature, including policy documents, technical standards, and reports from organizations such as World Health Organization (WHO), the World Bank, the H3Africa Consortium, Global Alliance for Genomics and Health (GA4GH), ISBER, Biobanking and BioMolecular Resources Research Infrastructure – European Research Infrastructure Consortium (BBMRI-ERIC), the African Union Commission, the Convention on Biological Diversity (Nagoya Protocol), and the FIND Biobank Network.

We built our search strings using Medical Subject Headings (MeSH) for PubMed, Emtree terms for Embase, and similar controlled vocabulary for Scopus and Web of Science. We also included free-text terms and combined them with Boolean operators. The full search strings for each database are provided in Supplementary file S1.

Two reviewers (AMF and KS) independently screened all retrieved records using Rayyan systematic review software (41). Title and abstract screening was conducted in a blind mode, in which Rayyan concealed each reviewer’s decisions until both had completed their screening of a given record. At the title and abstract stage, records were classified as include, exclude, or maybe. All records classified as maybe by either reviewer proceeded to full-text review. Disagreements between reviewers at both stages, defined as one reviewer selecting include, and the other selecting exclude, were resolved through synchronous discussion until consensus was reached. To be included, studies had to cover at least one of these topics: biobanking, genomic data management, digital health infrastructure, data governance, health informatics, precision medicine, or capacity-building in LMIC settings. Some studies from HICs were included if they provided important technical standards, theoretical frameworks, or other directly relevant ideas. Eligible publications included peer-reviewed research, systematic and scoping reviews, conceptual articles, policy analyses, consensus statements, official guidelines, and grey literature from recognized international organizations. Only English-language reports were included. We excluded reports that focused solely on biobanks in HICs without applicable lessons for LMICs, addressed only non-human biospecimens, were conference abstracts without an accompanying full report, or were opinion pieces without supporting evidence.

2.3 Data extraction, charting, and synthesis

We extracted data using a standard charting form (Supplementary file S2) that included: (1) author(s) and year, (2) country or region of study, (3) study design or document type, (4) main topic area, (5) key findings or arguments about the biospecimen-data divide, and (6) specific relevance to DIGS Model dimensions. Data were extracted into a standardized Microsoft Excel-based charting form (Supplementary file S2) by the lead reviewer (AMF). A random 25% sample of extracted records (n = 38) was independently verified by the second reviewer (KS) to assess extraction consistency. Any discrepancies identified in this verification sample were discussed and resolved by consensus before the verified extraction was accepted as final. We used Rayyan (41) to remove duplicate records across databases.

We analyzed the extracted data using thematic analysis following Braun and Clarke (42), with adjustments for literature-based evidence. We identified themes directly from the records, then mapped them to the four proposed DIGS Model dimensions. Our team went through several rounds of coding, developing themes, and refining concepts. The DIGS Model was developed from this process, not applied afterward. Its four dimensions were based on patterns of gaps and solutions found in the literature, and we formalized and named them during our research.

We used scoping review methods and did not formally assess the quality of individual studies, since our goal was to map concepts rather than combine effect estimates (38, 39). To check source credibility, we used a three-tier classification for all 154 studies. Tier 1 includes primary empirical evidence from peer-reviewed research conducted in, or directly relevant to, LMIC settings. Tier 2 covers policy and normative documents from major international organizations such as WHO, ISBER, GA4GH, ISO, and H3Africa. Tier 3 consists of conceptual, theoretical, and interpretive literature that provides analytical context. In the manuscript, we link claims to their evidence tier and clearly state when empirical claims are based only on Tier 3 records.

3 Results3.1 Search results and source selection

The systematic search found 2,390 records from five databases: PubMed/MEDLINE (620), Scopus (730), Web of Science (480), Embase (340), and Global Index Medicus (220). After removing 880 duplicates (36.82%), we screened 1,510 unique records by title and abstract. During title and abstract screening, we excluded 1,214 records because they did not focus on biobanking in LMICs, lacked relevant topics like digital infrastructure, interoperability, data governance, or skills, or did not meet the inclusion criteria. We reviewed 296 full reports for eligibility assessment. After full-report assessment, 154 studies from the database searches met all inclusion criteria. Figure 1 shows the PRISMA flow diagram (43).

PRISMA-ScR flow diagram for the DIGS Model scoping review. We identified records by searching five electronic databases from January 2010 to January 2026: PubMed/MEDLINE (620), Scopus (730), Web of Science Core Collection (480), Embase (340), and Global Index Medicus (220), for a total of 2,390 records. After removing 880 duplicates, 1,510 unique records were screened by title and abstract. We excluded 1,214 records because they were outside the topic, not focused on LMICs, or did not include a digital or data aspect. The final 154 studies included 59 original research articles, 36 narrative reviews, 22 perspectives and commentaries, 15 policy analyses and technical standards, 10 systematic and scoping reviews, and 6 primary grey literature documents. By DIGS dimension, there were 51 for Digital Infrastructure, 33 for Interoperability, 40 for Governance, 11 for Skills, 18 for background context, and 1 cross-cutting. Most of the evidence comes from sub-Saharan Africa, which is noted as a limitation in Section 6.2. GIM stands for Global Index Medicus; HIC for high-income country; LMIC for low- and middle-income country; and PRISMA-ScR for preferred reporting items for systematic reviews and meta-analyses extension for scoping reviews.

The 154 included studies comprised original research articles (n = 59), narrative reviews (n = 36), perspectives and commentaries (n = 22), policy analyses and technical standards (n = 15), systematic and scoping reviews (n = 10), and primary grey literature documents (n = 6). By topic, sources covered digital infrastructure (n = 51), governance (n = 40), interoperability (n = 33), skills (n = 11), background context (14), and one cross-cutting study, as summarized in Table 1.

CharacteristicCategoryn (%)Document typeOriginal research59 (38%)Narrative review36 (23%)Perspective/Commentary22 (14%)Policy analysis/Technical standard15 (10%)Systematic/Scoping review10 (7%)Grey literature6 (4%)Primary DIGS dimensionD1 Digital Infrastructure51 (33%)D2 Interoperability33 (21%)D3 Governance40 (26%)D4 Skills11 (7%)Background context18 (12%)Cross-cutting1 (1%)Geographic focusSub-Saharan Africa35 (23%)Global/Multi-region66 (43%)Latin America4 (3%)South/Southeast Asia7 (5%)Middle East/Eastern Europe5 (3%)HIC (comparator only)37 (24%)Evidence tierTier 1 (LMIC empirical)91 (59%)Tier 2 (Policy/normative)24 (16%)Tier 3 (Conceptual/theoretical)39 (25%)

Summary of included studies by document type, DIGS dimension, geographic focus, and evidence tier.

HIC-focused sources were included selectively as technical or conceptual comparators; they are not counted as primary LMIC evidence. Full extracted data for all 154 studies are provided in Supplementary file S2.

3.2 Deconstructing the biospecimen-data divide: four interconnected gaps

The biospecimen-data divide comprises several distinct yet interconnected gaps. Seeing these gaps as related, rather than as separate issues, is important for finding effective solutions. Figure 2 shows four main gaps: infrastructure, interoperability, human capital, and governance. Together, these form the foundation of the divide.

The biospecimen-data divide: four interconnected structural gaps in LMIC biobanking. The diagram shows four main gaps that together form the biospecimen-data divide in biobanking in low- and middle-income countries (LMICs). The infrastructure gap (top left) encompasses material problems such as unreliable electricity, poor connectivity, and insufficient computing hardware, all of which make it hard to generate data when collecting biospecimens. The interoperability gap (top right) refers to the fragmentation of health information systems that prevents them from sharing or using data, so biospecimens cannot be linked to ongoing clinical, genomic, and phenotypic records. The human capital gap (bottom left) refers to the shortage of people with the right mix of lab science, data management, bioinformatics, and research ethics skills needed to run digital biobanking systems. This is made worse by patterns that keep trained experts in high-income institutions. The governance gap (bottom right) points to the lack of strong national data protection laws, effective ethical review systems for genomic data sharing, and the use of international frameworks such as the Nagoya Protocol and the CARE principles for indigenous data governance. The arrows in the diagram show that all four gaps are connected: if one area is weak, it holds back the others. This is why single solutions do not fix the biospecimen-data divide. CARE stands for collective benefit, authority to control, responsibility, and ethics. LMICs means low- and middle-income countries.

3.3 The infrastructure gap: material conditions for data generation

The infrastructure gap refers to the physical conditions that help or hinder the collection of data from biospecimens. In many low- and middle-income countries, these conditions are often unreliable and vary by region (44). In many rural medical centers in sub-Saharan Africa and Southeast Asia, electricity can be unreliable or missing, making on-site data entry difficult (45, 46). Internet access, when available, is often too slow, costly, or unstable to transfer large genomic files, especially in areas that rely on mobile networks instead of fiber optics (47). Devices like tablets, laptops, and servers may also be in short supply, outdated, or not well-maintained (11, 15).

These physical challenges have many effects. When data entry at the collection site is not possible, staff must use paper forms, which can lead to mistakes and delays. Electronic health records, if available, often work as separate systems instead of being connected. This creates a divided data environment where biospecimens and their clinical data are kept in different, unconnected places (48). The infrastructure gap is more than simply a technical problem; it reflects more profound structural inequalities. The same global economic differences that place most biotechnology industries in HICs also affect how digital infrastructure is spread around the world (49). LMICs are expected to join a data-driven scientific field without the basic infrastructure that HICs already have (48, 50).

3.4 The interoperability gap: the problem of fragmented systems

Even when infrastructure is in place, another problem appears: different data systems often cannot communicate with each other. Interoperability, which means information systems can exchange and use data, is essential for biobanking to reach its full potential (51). Biospecimens are most valuable when they are connected to long-term clinical data, treatment results, and other health information (12, 52).

In LMICs, health information systems are frequently fragmented (53). One patient might have records in several separate systems, such as a tuberculosis registry, an HIV clinic database, a maternal health program, and a research biobank (54, 55). Even when unique patient identifiers exist, they may not be used consistently (56). Data standards also differ between systems, which makes integration difficult and time-consuming (57). Without shared standards like HL7 FHIR, LOINC, or SNOMED, exchanging data is still a manual and error-prone process (58–60). The BBMRI-ERIC provides a model for large-volume data harmonization. Its MIABIS (Minimum Information about Biobank Data Sharing) standard defines the core dataset needed to make biobank samples easy to find and use across institutions, and serves as a benchmark for measuring interoperability gaps in LMICs (61, 62).

This disintegration has both technical and epistemic dimensions. Technically, it requires investments in data standards, application programming interfaces (APIs), and system architecture (63). On the knowledge side, it provokes questions about which data are important enough to collect and connect. Deciding which data elements to standardize and whose standards to use is not a neutral process; it reflects specific scientific and political priorities (64). As a result, these systems shape what information is noticed and acted on in research communities (65). Also, when terminology and coding differ between platforms, analysts must spend extra time and resources matching data formats before they can combine information in a meaningful way (66).

3.5 The human capital gap: the missing cadre of data professionals

A third important gap is the need for skilled people inside digital infrastructure. Biobanking today needs staff who can combine lab skills for handling biospecimens, data science skills for managing multifaceted datasets, and ethical skills for issues like consent and governance (67). These “bio-data stewards” play a key role in turning biospecimens into useful knowledge (68). To develop this talent, organizations should create training programs that connect traditional lab work and data management (36, 69).

In LMICs, there are not enough of these professionals (70). Training has usually focused more on lab techniques than on data management, and career paths for data experts in health and research are often not well defined (71, 72). As a result, local teams commonly rely on outside partners for data analysis, which can create ongoing dependencies and reduce local control over research (34, 73). Missing clear career options, many trained staff leave for better opportunities in international health programs instead of staying to support local research (74).

This gap in skilled people matters because it affects who gets to ask questions about the data. When most data analysis happens in HICs, research commonly follows outside priorities instead of local health needs (75). Closing this gap is important for both technical progress and fairness in knowledge (76). If regional institutions build their own expertise, they can guide research and make sure health solutions fit their communities (77). To do this, local researchers need to be involved at every stage of research (78). Also, LMIC leaders should set research priorities so that data analysis assists local policy and health system improvements (79).

3.6 The governance gap: the ethical and regulatory vacuum

The fourth gap is about the governance frameworks needed to guide how data is collected, stored, shared, and used. In many LMICs, national data protection laws are either missing, outdated, or not properly enforced (80, 81). Across Africa, the African Union Data Policy Framework sets out principles for data sovereignty, cross-border data flows, and the management of health and genomic data among AU member states. This framework delivers a structure for national biobanking governance (82). However, most African LMICs have not fully put these continental policies into practice at the national or institutional level. For instance, there may be no specific laws on genomic data, and institutional review boards commonly lack genomics expertise. Ethical review committees may also be unfamiliar with the complex issues involved in genomic data sharing, dynamic consent, and international data transfer (83).

These issues are not simply theoretical. For example, the Havasupai Tribe case in the United States showed what can go wrong when blood samples collected for diabetes research were later used without permission for studies on schizophrenia and population migration (84). During the 2014–2016 West African Ebola outbreak, there were also disputes about exporting blood samples, demonstrating the ongoing conflict between quick pandemic response and fair data governance (85). When biospecimens were sent to labs in wealthier countries for analysis, people questioned whether the data and benefits would return to the communities and researchers who provided the samples (86).

The governance gap affects several levels. On the individual level, it can make consent meaningless, turning informed permission into just a formality (87). On the community level, it damages trust between communities and research institutions, which can discourage people from taking part in future biomedical research (88). On a larger scale, it keeps LMICs in the role of supplying raw materials, like biospecimens, instead of producing valuable results such as scientific discoveries, intellectual property, or commercial products (89–91). This disproportion is made worse by unclear rules about benefit sharing, which often do not protect researchers and participants in resource-limited settings from being exploited (92).

To close this gap, governance needs to go beyond just following rules and ticking checkboxes. Instead, it should focus on fairness, reciprocity, and shared control. Internationally, the Nagoya Protocol on Access and Benefit-Sharing sets legal requirements for countries and researchers who use genetic resources. It calls for prior informed consent, agreed terms, and fair sharing of benefits with the country where the resources come from. Even though the Nagoya Protocol is important for collecting and transferring biospecimens in LMICs, it is not well included in most biobanking governance frameworks there. Many international genomic research partnerships also do not clearly refer to its rules, which is a gap that the DIGS Model’s Governance dimension needs to address (93).

For governance in LMIC biobanking to be effective, it should be developed together with local partners, address community concerns, and follow principles like GA4GH and the CARE Principles (Collective Benefit, Authority to Control, Responsibility, Ethics) for Indigenous Data Governance. These steps help guarantee fairness and local control (94–96). Without these changes, the governance gap will keep harming both the ethics and the quality of genomic research within resource-limited settings.

3.7 The interconnections between gaps

These four gaps are connected and strengthen each other (97). Poor infrastructure makes it hard to set up interoperable systems, which leads to more segregated data silos and blocks smooth information sharing. This trend is clear when comparing different regions. Countries that invest more in digital systems, like Rwanda with its integrated health information system, show better interoperability than those with less developed infrastructure (98). Still, the situation is complicated. Good infrastructure by itself is not enough for interoperability without also investing in standards and workforce training (99).

When systems are not interoperable, it is harder to build local data expertise. Without standard technical frameworks, it is also difficult to provide practical training for a workforce skilled in advanced data management and analytics (34). A lack of local expertise makes it tough to set up and maintain strong governance, and weak governance in turn discourages investment in infrastructure (99). This creates a cycle where limited efforts to build capacity slow down national development plans and delay the growth of strong data ecosystems (100). The gap between biospecimen data and other data is made up of several related problems. To solve this, it is important to use universal, stable identifiers that establish lasting links between biospecimen catalogues and genomic databases (101). These technical procedures should be supported by clear institutional rules that make data curation a main priority, guaranteeing that trained and supported staff consistently follow standard workflows (102, 103).

3.8 Toward a new conceptualization: biobanking as an integrated digital Enterprise

The analysis above shows that current ideas about biobanking fall short. The main approach, often called the “physical repository model,” sees biobanks mostly as places to store biospecimens (36). In this view, success depends on how many samples are stored, how well they are preserved, and how efficiently they are distributed (104, 105). Digital systems, if considered at all, are seen as secondary rather than central to the biobank’s purpose.

In contrast, an integrated digital enterprise model focuses on the smooth movement of biospecimens and their related long-term data, turning static collections into active parts of a global research network (106). In this model, data processes are included from the beginning, and there is a focus on responding to new technologies. By using tools like artificial intelligence (AI) and blockchain with strong oversight, this approach tackles ongoing problems like tracking samples and keeping collections useful for science (26). In the age of precision medicine, this integration lets biobanks support high-speed sequencing, machine learning, and links to health records, which leads to insights at the population level (107). Each biospecimen is seen as a source of information, from DNA to physical traits. Making biobanking “FAIR”: making biospecimens and their data “findable, accessible, interoperable, and reusable”is part of this goal (108). This change means moving away from separate data management to systems that work together, making it easier to track samples and share data across institutions (26, 109). This adjustment also means using a “fit-for-purpose” quality approach, where careful documentation of each biospecimen’s history helps prepare for future technology (104).

Today, with genomics and precision medicine, a biobank that lacks digital systems cannot take part in, and may be left out of, the data-heavy research networks changing global health science (109). More importantly, the physical repository model hides the fact that biobanks are, and have always been, data organizations (110). By moving from a focus on samples to an emphasis on data, institutions can bring together different types of information, such as genomic, clinical, and pathology records (12). This change can help connect the gap between traditional biospecimen collection and the needs of precision health, where biospecimens are used for advanced analytics and insights at the population level (107).

3.9 Biospecimens as data carrier

Every biospecimen serves as a repository of critical data, encompassing the donor’s identity, the precise time and location of collection, relevant clinical context, and the biospecimen’s processing history. These data elements are not ancillary to the biospecimen; rather, they are key to its scientific significance. A biospecimen lacking such associated data cannot be distinguished from any generic biological tissue or blood sample (111). It is the presence of complete data that transforms a biospecimen into scientifically valuable evidence (112).

Recognizing biospecimens as fundamental data carriers has substantial impact on biobanking and biomedical resear