Kevin Yang

This paper develops a deep learning model called MolOR to predict odor percepts from odorant molecules by leveraging olfactory receptor activation profiles, showing improved prediction despite limited data overlap.

The paper examines the role of singleton protein sequences in training protein language models, finding they contain valuable functional diversity and should be included in training.

FLIP2 adds seven protein fitness datasets and 16 splits for enzymes, light-sensitive proteins, and interactions; simple ridge and zero-shot baselines often rival fine-tuned protein language models.

Vermeer generates synthetic microscopy images of protein localization using protein sequences and cell morphology, outperforming previous methods in quality and flexibility.

This paper introduces EnhancAR, a deep learning model that generates functionally diverse enhancers using evolutionary homologs instead of cell type labels.

This study introduces trainable subnetworks to analyze how protein language models factorize structural information and shows that structure prediction is sensitive to these factorized representations.

This paper introduces a pretraining task for genome language models that predicts evolutionary rate, showing improved performance over sequence-only pretraining.

CleaveNet is an AI pipeline that designs protease substrates with high efficiency and selectivity, validated experimentally for MMP13.

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.

The paper introduces the Dayhoff Atlas, a resource with large protein sequence datasets and generative models to improve protein design and prediction.

ProtNote is a multimodal deep learning model that enables supervised and zero-shot protein function prediction using free-form text inputs, generalizing to unseen functions.

PLAID generates controllable all-atom protein structures from sequences using a diffusion model trained on sequence data, enabling functional control through Gene Ontology and organism keywords.

This study analyzes transfer learning in protein language models, finding that downstream task performance often relies on early pretraining features rather than scaling with pretraining duration.

This paper introduces EvoDiff, a diffusion model that generates diverse, high-fidelity proteins in sequence space, overcoming limitations of structure-based models.

This paper introduces a self-supervised learning approach to detect and classify biosynthetic gene clusters in microbial genomes using masked language models trained on protein domains.

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.

The paper describes the evolution of iCn3D from a web-based 3D viewer to a batch-mode structural analysis tool with Node.js scripts for large-scale structural dataset analysis.

This paper introduces evolutionary velocity, a method using protein language models to predict evolutionary order and direction, showing its effectiveness across different timescales and providing new insights into viral-host immune escape and horizontal gene transfer.