Pengyu Ni

This paper develops models linking histological images to genotype, transcriptome, and age, identifying 906 imageQTLs and predicting gene expression and chronological age from images.

The paper expands and refines a reference catalog of human non-canonical open reading frames (ncORFs) by incorporating more datasets and introducing a translation signature score framework to prioritize high-quality ncORFs.

The paper expands the lncRNA catalog using capture long-read RNA sequencing, significantly increasing the number of annotated transcripts in human and mouse genomes.

The paper profiles the SHANK3 transcriptome using cDNA-capture and long-read sequencing, revealing complex transcript variants altered in Shank3 mutants and ASD brains, improving variant detection in neuropsychiatric disorders.

The paper processes single-nuclei multi-omics data from 388 human brains to identify regulatory elements and build cell-type networks, revealing changes in aging and neuropsychiatric disorders.

The paper predicts functional states and types of cis-regulatory modules (CRMs) across 107 cell/tissue types, revealing that dual-functional CRMs and silencers are more prevalent than previously thought.

This paper fine-tunes a large language model to predict protein phase transitions from sequence data and shows its utility in assessing sequence variants' effects on aggregation and droplet formation.

The paper identifies prevalent false positives and false negatives in STARR-seq data and explores their underlying causes using predicted cis-regulatory modules and cell line data.

The paper creates a comprehensive map of cis-regulatory modules and transcription factor binding sites in 80% of the mouse genome using ChIP-seq data.

The paper predicts functional states of cis-regulatory modules in humans and mice using epigenetic marks and machine learning, finding common epigenetic rules across species.

PCRMS provides a database of predicted cis-regulatory modules and transcription factor binding sites in human and mouse genomes, enabling exploration and analysis of regulatory sequences.

The paper predicts cis-regulatory modules in the human genome, finding they are more prevalent than previously thought.

The paper uses deep learning to identify sequence determinants of histone modification patterns in CD4+ T cell differentiation.

ProSampler efficiently identifies transcription factor binding motifs in large ChIP-seq datasets, revealing complex motif patterns and cooperative binding mechanisms.