Jianfeng Sun

memVar links human membrane protein variants to molecular contexts and disease, revealing enrichment of pathogenic variants in transmembrane regions and interaction sites.

CHORUS predicts membrane protein interaction interfaces from sequence, outperforming existing methods and providing a proteome-scale atlas of interactions.

This study compares nanopore direct RNA and PCR-cDNA sequencing protocols, showing distinct 5' and 3' coverage biases affecting transcript-level analysis but similar gene-level accuracy.

UMIche provides a unified platform for UMI-centric simulation and analysis in bulk and single-cell sequencing to improve molecular quantification accuracy.

Tresor simulates transcriptomic reads with PCR error representation across sequencing technologies, showing cell-type-specific profiles similar to reference data after duplicate removal.

mclUMI improves UMI deduplication by using Markov clustering to dynamically remove PCR duplicates, outperforming existing methods in noisy sequencing conditions.

The paper introduces an anchor-enhanced bead design to reduce oligonucleotide synthesis errors in single-cell sequencing, improving UMI recovery and feature detection.

The paper demonstrates that PCR errors significantly affect UMI-based absolute counting in sequencing data and proposes a solution using homotrimeric nucleotide blocks to correct these errors.

DeepsmirUD predicts small molecule-mediated regulatory effects on miRNA expression using deep learning, achieving high accuracy and enabling drug repositioning insights.