AI Emerges as a Research Collaborator in Biology, Opening New Possibilities for Medical Science

Artificial intelligence (AI) is increasingly becoming a research collaborator in biological sciences, helping scientists analyse complex datasets, develop hypotheses and explore new possibilities in healthcare and medicine. Recent research involving Anthropic’s AI system Claude, which helped identify a previously unknown enzyme system in viruses that infect bacteria, highlights the growing role of AI in scientific discovery.

The enzyme system, called array-associated reverse transcriptases (ART), was identified in bacteriophages, viruses that infect bacteria. According to Anthropic, the AI-assisted discovery took approximately 21 hours, a process that could otherwise require researchers weeks or months of work.

This development signals a shift from using AI merely to process scientific information towards employing it as an active participant in research. However, scientists emphasise that human creativity, critical thinking, experimentation and oversight remain essential to validate AI-generated findings.

AI Applications in Genetic Research, Protein Structures and Drug Development

AI tools are helping researchers examine large biological datasets, including DNA sequences, protein structures and medical imaging. Platforms such as NotebookLM, ResearchRabbit and Elicit also assist with literature reviews, research planning, citation tracking and scientific analysis.

Google DeepMind’s AlphaGenome can analyse DNA sequences of up to one million base pairs to predict how genetic variations may influence gene regulation. Such capabilities could help scientists investigate disease mechanisms and understand genetic variations, particularly in rare diseases where identifying the effects of changes in non-coding DNA regions remains challenging.

Another significant development is AlphaFold2, an AI system that predicts protein structures from amino acid sequences. Its contribution to protein-structure prediction was recognised through the 2024 Nobel Prize in Chemistry awarded to Demis Hassabis and John Jumper, alongside David Baker for computational protein design.

Understanding protein structures helps researchers investigate diseases, identify potential drug targets, develop vaccines and explore existing medicines for new uses. AI-based predictions can also reduce the time and resources required for preliminary research before laboratory testing.

From Predicting Biology to Designing New Biological Systems

Generative AI is taking biological research further by enabling scientists to design genetic sequences and biological systems that do not naturally exist. Researchers at Stanford University and the Arc Institute have used the Evo 2 model to design bacteriophage genomes capable of targeting antibiotic-resistant Escherichia coli strains. Laboratory testing of AI-designed genomes demonstrates the potential of these tools to support research into alternatives to conventional antibiotics.

AI is also influencing drug discovery. Rentosertib, a drug candidate developed by Insilico Medicine using AI-driven discovery methods, entered Phase III clinical trials, marking progress towards integrating AI into pharmaceutical development.

Despite these advances, AI-generated hypotheses and biological designs require rigorous experimental validation. Researchers must also address questions of transparency, accountability, safety and the potential misuse of biological technologies.

The future of AI-enabled biology will therefore depend not only on computational power but also on responsible scientific practices. By combining machine-driven analysis with human judgement and laboratory verification, researchers can explore new discoveries while maintaining scientific reliability and ethical accountability.