Jian Zhan

The paper assesses the completeness of RNA fragment structural space and finds it incomplete, especially at higher oligomer levels, with fewer backbone fragments compared to proteins.

ESM-LoRA-Gly improves prediction of N- and O-linked glycosylation sites by fine-tuning ESM2-3B with LoRA, achieving better accuracy and efficiency.

RESM improves RNA sequence-structure-function prediction by transferring knowledge from protein language models, outperforming existing models in multiple tasks.

The paper designs a protein that binds to multiple SARS-CoV-2 RBD variants with high affinity, demonstrating a strategy for broad-spectrum antiviral inhibitors.

RPcontact predicts RNA-protein contacts using RNA and protein language models, outperforming existing methods on contact prediction.

The paper introduces MobiSeq, a method combining deep mutational scanning and mobility-based selection to predict RNA tertiary structures by analyzing covariational signals and improving energy-based predictions.

This paper introduces SPIN-CGNN, a protein design method using contact maps and graph neural networks, showing improved sequence recovery and structural accuracy compared to existing methods.

The paper shows that using artificially generated, structure-stability-selected homologous sequences improves AlphaFold2's accuracy for protein structure prediction, achieving better than 2 Å RMSD for most tested proteins.

This paper introduces an RNA language model based on multiple sequence alignment, showing it can accurately predict RNA structure from sequence data.

This study evaluates deep learning models for RNA base-pairing prediction, showing that SPOT-RNA can accurately predict RNA-RNA interactions in unseen structures and without monomeric structures.

The paper builds a large RNA sequence database (MARS) and integrates it with RNAcmap to improve homology search and RNA structure/function inference.

RNAcmap automatically predicts RNA contact maps using evolutionary coupling analysis, outperforming manual methods and showing consistency across RNA sequences.

The paper identifies a self-cleaving ribozyme in ancient LINE-1 elements, showing its emergence 40 million years ago and extinction 30 million years ago.

The paper uses deep mutational scanning to accurately infer the full base-pairing structure of RNA with high precision.

The paper develops a method to distinguish experimentally validated lncRNAs from high-throughput lncRNAs and mRNAs using Support Vector Machines, achieving high accuracy in human and mouse data.