Chaired by Professor Muzlifah Haniffa and Professor Ben Lehner

Overview

The goal of this workshop was to develop a long-term vision for understanding and engineering living systems, focusing on the emergent properties that govern biological complexity and exploring how genomes and cells might be harnessed to address challenges beyond traditional biological contexts.

Central to discussions was the vision of harnessing advancements in artificial intelligence (AI) and synthetic biology to enable predictive, personalised, and ethically grounded healthcare. These technologies can extend benefits from individual patients to population-level impact, transforming both disease prevention and intervention.

While tools such as the AlphaFold protein structure database demonstrate the value of standardised data for predictive biology, broader challenges persist due to limited integrative datasets and the complexity of synthesising and testing diverse protein combinations. These challenges illustrate why imagining futures, rather than simply predicting outcomes, is important for genomics as a foundational capability across disciplines.

Framing the Challenge: Genomics Beyond the Next Grant Cycle

Genomics is increasingly positioned as a foundational capability across scientific disciplines. Short-term, project-based approaches are insufficient for addressing long-term questions about biology and health.

Workshop discussions highlighted the need for long-horizon thinking that connects immediate research outputs to transformative impacts on society, medicine, and our understanding of life itself.

Advances in AI and computational modelling allow the prediction of molecular and cellular behaviour at unprecedented scales, but these predictive tools raise questions about understanding. Can we claim to understand biological systems if we can predict outcomes without explaining mechanisms?

Participants emphasised that foundational, hypothesis-generating research is essential to answer the fundamental questions of what, why, and how.

What Does It Mean to Understand Life in the Age of AI?

Workshop participants discussed three complementary dimensions of biological research:

• prediction

• explanation

• engineering.

AI can generate highly accurate predictive models, but these may lack mechanistic insight. Combining predictive power with experimental perturbation, including synthetic biology, allows causal relationships to be explored and emergent behaviours to be studied.

The duality of ‘day science’ and ‘night science’ was highlighted. ‘Day science’ refers to hypothesis-driven, precise, disciplinary approaches, while ‘night science’ is exploratory, interdisciplinary, and metaphorical. Both are required to generate transformative scientific insights. AI may accelerate night science by generating hypotheses at scale, but human judgement, creativity, and interpretation remain essential for guiding experiments and maintaining robust research culture.

Building Biology on Demand: Promises and Limits of Engineering Life

Genome and cell engineering are emerging as general-purpose tools for manipulating living systems. Participants explored two complementary approaches:

Natural modification versus de novo synthesis

Some applications focus on editing existing genomes, while others aim to construct entirely new sequences using synthetic nucleotides. Both approaches face challenges of reproducibility, control, and robustness and require careful consideration of societal permission, safety, and governance.

10-Year Vision 1: Engineering Controllable Biological Systems

Deconstructing and Reconstructing the Units of Life: Synthetic Blood

This vision describes a future in which synthetic immune system models based on human haematopoietic stem cells (HSCs) can generate any blood cell type on demand. The goal is to create controllable platforms to study immune responses, blood cell development, and disease susceptibility.

Mechanistic insights would be pursued through a combination of synthetic biology, computational modelling, and targeted genetic perturbations. By systematically altering regulatory components, researchers aim to map differentiation pathways and identify factors that govern functional specialisation.

Key challenges include bias in existing datasets, particularly a lack of ancestral and environmental diversity, which limits predictive model generalisability. In addition, incomplete knowledge of cell differentiation and essential transcription factors constrains model reliability.

Experimental platforms, including in vitro microfluidics and whole-organism models such as zebrafish, will be essential for validating differentiation pathways and observing developmental dynamics in vivo. This approach highlights the intersection of mechanistic investigation, technological innovation, and rigorous experimental validation.

10-Year Vision 2: Building Biology on Demand at Scale

Engineering by Natural Modification versus de novo from Synthetic Nucleotides

This vision describes a future in which scalable, cost-effective genome engineering and interoperable data ecosystems allow biology to be designed, tested, and deployed widely, contingent on public legitimacy and sustained investment.

Major advances would enable synthesis and editing of long DNA sequences at any scale and reduce genome synthesis costs by two orders of magnitude. Synthetic biology would be integrated into society through demonstrable public benefits, including health and ecological applications.

The vision also explores non-DNA-based innovations and alternative biological systems to expand the conceptual and technical foundations of biology. Democratising access to computational tools, data, and AI models is central to this approach. Standardised formats and cross-platform interoperability would be supported by institutions and existing funding infrastructure.

Challenges include public scepticism of synthetic biology and fragmentation across data repositories. Addressing these barriers is critical to scaling capability while maintaining public trust and enabling equitable participation.

From Precision Medicine to Precision Global Health

Technological innovation underpins the emerging concept of precision global health, extending the promise of highly individualised care to population-level impact. Key drivers include:

Precision global health relies on interdisciplinary collaboration and data integration, enabling actionable insights across populations while maintaining attention to ethical, societal, and environmental considerations.

Data, Models, and Bottlenecks in Progress

Participants identified data generation and integration as major constraints on discovery.

10-Year Vision 3: Learning Generalisable Rules of Living Systems

Modelling Multiscale Interactions and Emergent Properties

This vision focuses on deriving generalisable rules governing emergent properties in simple multicellular systems to support predictive models. Research would progress in a structured manner, moving from cell co-cultures to 2D bioprinted models, 3D organoids, organs, and multi-organism interactions.

Cellular communication and cross-species interactions, inspired by natural collective systems such as bird flocking and bee cooperation, would inform the study of emergent behaviours.

Success depends on fostering global interdisciplinary research communities integrating biology, ecology, and social sciences alongside continued development of organoid culture and bioprinting technologies. Collaboration across disciplines is essential for addressing the complexity of multiscale biological systems.

Who Does Science and Who Benefits

Discussions highlighted power imbalances and inequities in the global research ecosystem.

Equitable engagement is not just a moral imperative; it also enhances innovation, improves dataset diversity, and strengthens the relevance of discoveries to global populations.

Public Trust, Ethics, and Societal Permission

Scientific advancement cannot be separated from societal values and ethics:

Ethical and societal considerations must evolve alongside technical advances, ensuring benefits are safe, equitable, and socially accepted.

Looking to 2050: Questions Beyond the Next Decade

Beyond the 10-year horizon, several long-term considerations emerged:

Open Questions and Directions of Travel

Key unresolved questions highlight tensions between ambition, ethics, and feasibility:

Together, these questions frame a long-term research agenda that integrates technical, societal, and ethical dimensions, highlighting that the future of genomics depends as much on governance and culture as on experimental capability.

Challenges and Considerations for Future Health Research

Demographic changes emerged as a significant driver of future health challenges. Rapid population growth in Africa and ageing populations in Europe increase genetic diversity and drive a shift from communicable to non-communicable diseases. Addressing these trends will require targeted interventions informed by network theory, alongside technological enablers such as AI, quantum computing, and equitable international collaborations.

Understanding biology in this context raises deeper scientific questions. Can predictive models alone provide genuine understanding, or must mechanisms be explained? Workshop discussions emphasised the need for foundational, exploratory science, adopting a “slow science” approach to answer fundamental questions. The duality of day science (hypothesis-driven and precise) and night science (exploratory and interdisciplinary) remains critical. While AI may accelerate hypothesis generation, the human element is indispensable for interpretation, judgement, and fostering research culture.