The Future of Biobanking: Integrating AI and Omics for Smarter Public Health

In a comprehensive review, researchers Pedro Venturini, Paula Lobato Faria, and João V. Cordeiro from the NOVA National School of Public Health and the Interdisciplinary Center of Social Sciences at NOVA University Lisbon explore the fast-evolving convergence of artificial intelligence (AI), omics science, and biobanking. Conducting a structured literature review of 37 significant studies, the authors examine how these fields are reshaping public health (PH) research and practice, while also analyzing the technical, ethical, and regulatory hurdles that must be overcome to ensure responsible and equitable implementation.

Biobanks, as curated repositories of biological samples and associated metadata, including genomic, lifestyle, and environmental data, are increasingly pivotal in modern medical and public health systems. Large-scale projects like the UK Biobank, the NIH’s “All of Us” program, and the European “1+ Million Genomes” initiative exemplify global efforts to integrate biobanking into national health strategies. These biorepositories not only facilitate precision diagnostics and treatments but also enable population-scale research into disease trends, health disparities, and gene-environment interactions. In this context, AI and omics technologies offer tools to manage, analyze, and extract meaningful insights from this growing ocean of data.

Transforming Data into Health Solutions with AI

Artificial intelligence is fast becoming an indispensable companion to modern health data science. The article outlines AI’s evolution from task-specific Artificial Narrow Intelligence (ANI) to theoretical constructs like Artificial General Intelligence (AGI) and Artificial Superintelligence (ASI). Particularly impactful are recent advances in Generative AI (GenAI) and multimodal AI, which allow machines to simultaneously process text, images, and sound, similar to how clinicians interpret multiple forms of input when diagnosing patients.

These advances are proving essential in analyzing omics data, an umbrella term that includes genomics, transcriptomics, proteomics, metabolomics, and exposomics. AI’s ability to handle nonlinear, high-dimensional datasets makes it especially suitable for uncovering molecular patterns that underlie complex diseases. Through deep learning and pattern recognition, AI systems can identify genetic mutations, discover biomarkers, and support the development of precision therapies. One breakthrough example is AlphaFold, an AI tool that now predicts protein structures with an accuracy previously limited to experimental methods, revolutionizing proteomics and drug discovery.

High-Throughput Sequencing and Precision Medicine

One of the most transformative intersections of AI and omics lies in high-throughput sequencing (HTS). By reducing sequencing costs and increasing data generation exponentially, HTS has empowered scientists to analyze entire genomes with remarkable speed and precision. AI enhances these processes by automating data analysis, improving interpretability, and enabling the integration of diverse omics layers. This synergy has played a vital role in cancer genomics, rare disease diagnostics, and even infectious disease surveillance, notably during the COVID-19 pandemic.

In addition to diagnostics, HTS also supports gene editing technologies like CRISPR/Cas9. While gene editing has existed since the 1980s, recent developments have dramatically improved its precision, with AI playing a central role in designing and validating new gene-editing tools. By minimizing off-target effects and enhancing delivery mechanisms, AI-powered gene editing opens new avenues for treating both inherited and complex diseases. These innovations are making it possible to move from bench to bedside faster than ever before, with significant public health implications.

Opportunities and Risks in a Data-Driven Health Future

Despite these opportunities, the article doesn’t shy away from the significant challenges posed by integrating AI and omics into biobanking. Technical issues include model selection, algorithm validation, data standardization, and a pressing need for explainability in AI outputs. Large models often operate as “black boxes,” making their decisions difficult to interpret, an unacceptable risk in clinical or policy contexts. Ethical concerns are equally prominent, particularly around privacy, data ownership, bias, discrimination, and transparency.

The authors warn of growing disparities if governance fails to keep pace with technological advancements. In worst-case scenarios, private corporations may monopolize biobank infrastructures, reducing government oversight and increasing inequalities in healthcare access. To mitigate these risks, the authors advocate for inclusive governance models that promote transparency, protect individual rights, and engage diverse stakeholders. This includes adopting robust regulatory frameworks that ensure accountability, validation standards, and continuous monitoring of AI systems.

Four Futures for Public Health: From Hope to Alarm

To visualize potential outcomes, the authors present four plausible future scenarios: Gradual Optimism, Disruptive Pessimism, Contingent Optimism, and Pessimistic Social Shaping. These are based on how technology, governance, and society might interact. The Gradual Optimism scenario imagines equitable global access to AI-powered diagnostics and international cooperation on governance. In contrast, Disruptive Pessimism foresees a world dominated by technological determinism, where health data is controlled by unregulated private entities and public institutions are sidelined.

Contingent Optimism offers a balanced path forward, envisioning political leadership that adapts regulation to ensure ethical, transparent, and equitable deployment of AI in PH. Finally, Pessimistic Social Shaping presents a fractured health ecosystem, stratified by socio-economic status and controlled by data monopolies. These scenarios are not predictions but serve as tools to guide policymakers, technologists, and civil society in shaping a healthier, fairer future.

The authors emphasize that the integration of AI and omics into biobanking holds immense promise for improving population health, but its success hinges on equitable access, robust data governance, and sustained interdisciplinary collaboration. Ensuring that technological advances serve public interests, rather than deepening existing divides, will be critical for realizing the full potential of these transformative tools in public health.