Pan Li

Harnessing AI to Build Virtual Cells

This paper introduces VCHarness, an AI system that automatically builds perturbation-response models for virtual cells, outperforming manual methods and reducing development time.

Training Compute-Optimal Protein Language Models

This paper investigates optimal training of protein language models by analyzing the relationship between model size, training data, and performance, finding that diverse training data and scaling laws improve results.

An end-to-end approach for protein folding by integrating Cryo-EM maps and sequence evolution

This study introduces SMARTFold, a deep learning model that integrates Cryo-EM maps and sequence alignment to predict protein structures, outperforming existing methods on benchmark datasets.

Differential Analysis of RNA Structure Probing Experiments at Nucleotide Resolution: Uncovering Regulatory Functions of RNA Structure

DiffScan identifies structurally variable regions in RNA structure probing data with improved accuracy across multiple platforms.

In vivo structural characterization of the whole SARS-CoV-2 RNA genome identifies host cell target proteins vulnerable to re-purposed drugs

The paper maps the in vivo structure of SARS-CoV-2 RNA, identifies novel structural features, and predicts host proteins interacting with the RNA using deep learning.

U1 snRNP regulates chromatin retention of noncoding RNAs

The paper identifies a U1 snRNP recognition motif that regulates chromatin retention of noncoding RNAs through experimental and computational methods.

Dynamic Regulation of RNA Structure in Mammalian Cells

The paper maps RNA structuromes in three subcellular compartments and identifies LIN28A as an m6A anti-reader, revealing dynamic RNA structure regulation.

Integrative analysis of Zika virus genome RNA structure reveals critical determinants of viral infectivity

This study identifies RNA structural elements in ZIKV strains that influence infectivity through comparative in vivo analysis.