Research & Development
PhD scholarship in generative protein and peptide design for sustainable biomaterials - BRIGHT
Summary
Develop computational methods for designing protein- and peptide-based biomaterials, combining generative models with iterative laboratory feedback. The role includes collaboration with experimentalists, research dissemination, and teaching and supervision.
The fields are read by AI from the ad. If the employer filled in a field themselves, we use theirs.
Opportunity summary
This PhD project is part of a BRIGHT programme developing a general platform for the design of protein- and peptide-based biomaterials, with solid-state battery components as its first testbed. Our goal is to replace fluorinated PFAS polymers with de novo designed proteins and peptides.
You will generate the computational designs at the heart of the platform, by combining molecular foundation models with iterative laboratory feedback. You will develop design algorithms for biomaterials with natural and non-natural amino acids, and for planning their biosynthesis and biomanufacturing.
You will develop lab-in-the-loop algorithms that use generative model-controlled stochastic synthesis to maximize information gain at every round of experimental testing.
Responsibilities and qualifications
Responsibilities
- Generate de novo peptide and protein designs meeting multiple coupled property targets, including adhesion, cohesion, ion conduction and controlled degradability
- Use and adapt generative molecular design tools such as RFdiffusion, ProteinMPNN and AlphaFold-type models
- Develop and extend information-dense synthesis algorithms to manufacture designs in the laboratory
- Extend these methods to expanded genetic codes, so that non-canonical residues can be encoded and biosynthesised rather than only chemically synthesised
- Develop methods to learn from diverse measurements of material properties, via shared latent representations
- Develop methods to continuously learn from iterative experimentation, by feeding back experimental data into an integrated model with quantified uncertainty
- Work closely with experimentalists to optimize experimental designs for collecting high-information training data
- Work closely with experimentalists to automate and scale up design-build-test-learn cycles
- Report your research results in peer-reviewed scientific publications and international conferences
- Collaborate within the research group and our local, national and international networks
This is an extract. Read the whole advert at Employer's career site