(127a) Fast Single-Sequence Protein Structure Prediction with Applications in Protein Design and Novel Biomaterials

Recent advances in protein modeling (specifically, AlphaFold2 from Google, DeepMind) have made it possible to predict protein structures with high fidelity from alignments of homologous protein sequences by using significant computational resources. While groundbreaking, three outstanding challenges remain unaddressed by these systems: (i) prediction of structure from individual sequences, necessary for orphan proteins, de novo design, rapidly evolving proteins, and modeling genetic variation, (ii) fast prediction, necessary for protein design and whole-proteome analyses, and (iii) scientific understanding of the sequence-to-structure relationships that underpin protein folding.

To this end we (a) predicting protein structure from single protein sequences in an end-to-end differentiable manner without usage of any explicit form of evolutionary information to provide an alternate route to accurate and fast protein structure prediction, and (b) create extensive structure-to-function mapped libraries to enable a genomics-equivalent paradigm called ‘struct-omics’ – with focused applications in (a) protein design, (b) enzyme engineering, (c) porin-based separation devices, and (d) novel biomaterials. My brand of research will create strong collaborations across disciplines of computer science, chemical engineering and biomedical sciences and will utilize the power of machine learning and optimization to explore hitherto unknown biology.