Research
For billions of years, nature has been conducting the greatest experiment of all time. Imagine one-day gaining access to the detailed notes from these experiments. Today, with worldwide expeditions to collect samples from all habitats, single-cell sequencing of unculturable microbes and the rapid drop in sequencing costs, we can finally tap into nature and gain access to these notes.
Making use of this data, our lab is interested in:
Developing a unified statistical model of protein evolution that integrates phylogenetics, genomic, structural, and functional constraints.
Recent Publications
Hettiarachchi R, et al. 2023. πGithub
Differentiable Search of Evolutionary Trees from Leaves.ΒHwang Y, et al. 2023. πGithub
Genomic language model predicts protein co-regulation and function.ΒPetti S, et al. 2022. πGitHub
End-to-end learning of multiple sequence alignments with differentiable Smith-Waterman.Β
Zhang Z, Wayment-Steele HK, et al. 2024. πGithub
Protein language models learn evolutionary statistics of interacting sequence motifs.ΒWang H, et al. 2022.πGithub
Disentanglement of Entropy and Coevolution using Spectral Regularization.ΒBhattacharya N, et al. 2021. πGithub
Interpreting Potts and Transformer Protein Models Through the Lens of Simplified Attention.ΒMarshall D, et al. 2020. πGithub
The structure-fitness landscape of pairwise relations in generative sequence models.Dauparas J, et al. 2019. πGithub
Unified framework for modeling multivariate distributions in biological sequences.Β
Explicit modeling of the protein conformational (and/or folding) landscape for protein structure prediction and design.
Recent Publications
Cho Y, et al. 2024. πGithub
Implicit modeling of the conformational landscape and sequence allows scoring and generation of stable proteins.Roney JP, et al. 2022. πGithub
State-of-the-Art estimation of protein model accuracy using AlphaFold.ΒOvchinnikov S, et al. 2021.
Structure-based protein design with deep learning.ΒNorn C, et al. 2021. πGithub
Protein sequence design by conformational landscape optimization.
(image credit Basile Wicky)Β
Applying the models to mine metagenomic βdark matterβ sequences to discover new protein families, functions, and protein-protein interactions. Probing evolution of multicellularity through comparison of structures and interactions in the early tree of life.
Recent Publications
Pavlopoulos GA, et al. 2023.πDatabase
Unraveling the functional dark matter through global metagenomics.ΒTrinquier J, et al. 2022.πGithub
SWAMPNN: End-to-end protein structures alignment.Ovchinnikov S, et al., 2017.πGithub
Protein structure determination using metagenome sequence data.Β
All Publications