EvoDiff
A diffusion framework for controllable protein sequence and evolutionary alignment generation using discrete diffusion models.
What is EvoDiff?
EvoDiff is an open-source diffusion framework for generating protein sequences and evolutionary alignments using discrete diffusion models. It enables controllable protein design by combining evolutionary-scale data with diffusion-based generation, allowing researchers to create novel proteins with desired functional or structural properties. The framework addresses limitations of structure-based models by operating directly in sequence space, making it possible to design proteins with intrinsically disordered regions.
Computational biologists, protein engineers, and bioinformatics researchers who need to generate novel protein sequences or design proteins with specific functional motifs, disordered regions, or evolutionary constraints.
EvoDiff provides a flexible, sequence-first approach to protein design that goes beyond traditional structure-based methods, offering both unconditional and conditional generation capabilities. Its unique value lies in leveraging evolutionary information through MSAs and enabling the design of proteins with intrinsically disordered regions, which are inaccessible to most existing protein design tools.
Overview
Generation of protein sequences and evolutionary alignments via discrete diffusion models
Use Cases
Best For
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
