Fajie Yuan

This paper introduces Pinal, a foundation model that designs proteins from natural language descriptions, achieving functional alignment and experimental success in four design tasks.

This study designs epitope-focused immunogens targeting conserved coronavirus regions, showing broad neutralizing activity and protection in mice against diverse coronaviruses.

DOMINO generates multidomain proteins by learning domain co-occurrence patterns and using them to design novel sequences with high structural confidence.

The paper introduces a foundation model and automated pipeline for cryo-EM image processing, improving efficiency and quality assessment in structural biology.

The paper introduces SaprotHub, a platform that simplifies training and sharing of protein language models, enabling non-experts to build and collaborate on customized models.

ProTrek introduces a tri-modal contrastive learning model for protein sequence, structure, and function, achieving significant improvements in speed and accuracy over existing tools.

PRIME is a multi-agent system that dynamically synthesizes computational workflows for protein engineering, outperforming existing AI agents in complex tasks.

Evolla, a protein-language model, improves functional inference and discovers novel proteins through natural language queries and benchmarking.

This study uses protein language models to engineer a uracil-N-glycosylase variant for programmable T-to-G and T-to-C base editing, achieving high success rates in correcting genetic mutations in mice.

Pairing each residue with its Foldseek 3D-structure token and pre-training an ESM-2-sized model on about 40 million AlphaFold2 structures beat ESM-2 and structure-based baselines on variant-effect and function tasks.

This paper introduces a diffusion model-based pipeline for antibody design and optimization, achieving high performance in structure and sequence design and docking, validated by improved binding to flaviviruses.

The paper compares sequence-based and structure-based methods for predicting protein function, finding that structure-based approaches can offer better accuracy when using high-quality protein structures.

FunGeneTyper is a deep learning framework for ultra-accurate classification and discovery of functional genes in microbiomes, outperforming existing methods in multiple environments.

This study shows that pre-trained protein language models can predict mutation pathogenicity and patient survival in cancer, outperforming existing methods.