AI models that generate images from text prompts now face a harder test: designing proteins that actually work in a lab. Current generative models for biomolecules produce thousands of candidates, but fewer than one in a hundred survive experimental testing. This project tackles that failure rate head-on. The researcher will build new computational methods that model molecules not as flat images but as three-dimensional objects with physical and chemical constraints. Instead of copying techniques from image generation—which ignore how atoms bond and fold—the work will develop architectures tailored to molecular geometry. If successful, the approach could slash the number of dead-end designs that waste lab time and materials. That would accelerate the development of new enzymes for industrial catalysis, therapeutic proteins, and synthetic biology tools. The project also aims to give scientists precise control over properties like stability or binding affinity, so they can dial in a desired function rather than sifting through random outputs. This is fundamental computational science. It will not produce a market-ready drug or catalyst. But by improving how machines represent and generate molecular structures, it could make the entire pipeline from computer model to working protein faster, cheaper, and far more reliable.
View original technical description
Generative modelling techniques like diffusion and flow matching have revolutionised fields like image generation and are about to do the same in the scientific domain, particularly in synthetic biology where they are used to generate novel molecules useful as therapeutics, catalysts and many more applications. However, the success rates of these models remain very low, and their controllability remains weak, reducing their practical impact. This project aims to create powerful and practically useful generative models of biomolecules for applications in protein design and beyond. The main question this project intends to answer is What is the most efficient way to model and design a biomolecule computationally so that it has a high chance of fulfilling the design constraints in the real world? Efficiency here involves both computational efficiency in terms of compute resources such as time and memory, but also experimental efficiency in terms of the success rate these designed molecules show once they are manufactured and tested in the laboratory. Major themes that will be explored are 1. Learning/modelling molecular representations that are both computationally efficient as well as performant in the task at hand; this will involve both representation in terms of granularity (do we model a molecule atom-by-atom or in a coarse-grained fashion?) as well as in terms of geometrical invariances (do we learn necessary equivariances or do we enforce them by architecture constraints?) 2. Improving fundamental neural network architectures and training procedures for generative modelling on biomolecules. Many current approaches are directly transferred from image generation, and initial investigations show that these are often suboptimal. 3. Enhancing controllability and raw distribution modelling capacity of generative models. In generative modelling for other modalities like images, metadata like image captions are used to subdivide the hard general distribution modelling problem into many smaller ones. Doing something similar in the domain of biomolecules promises to improve performance and allow practitioners to use these models a lot more effectively and tuned towards their needs. 4. Including experimental constraints: in many cases, downstream laboratory experiments have practical constraints that are important to consider in model design, whether that is as part of which data is available as input to the model or what are the experimental limitations in terms of synthesizing the predicted molecules. Current models often fail to be integrated into a rapid feedback loop by neglecting these considerations, including them into model design promises to make these models more impactful in practice. 5. Improving benchmarking and evaluation of models: Since the use of generative models in biology started only recently, it is not fully clear how to evaluate in practice which model is the best, especially given the limited experimental evaluation that is performed (see point above). Therefore, one focus of this thesis will be developing better in-silico benchmarking metrics and tools to guide model development in a more streamlined way. This project falls within the EPSRC "healthcare technologies" research area and will be carried out with collaborators in Oxford (such as Professor Charlotte Deane) as well as collaborators abroad.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know