Jun S. Song

Chromosome 3p deletion leads to extensive genomic alterations in diverse cancers and confers synthetic lethality in uveal melanoma

This study identifies chr3p deletion as a driver of genomic instability in cancers, showing synthetic lethality in uveal melanoma through MITF disruption and feedback loops with MYC and GNAQ/GNA11.

UV induces common cutaneous amyloid-like melanosomal protein aggregates

The paper shows UV exposure causes misfolded melanosomal proteins to aggregate, contributing to skin pigmentation disorders and melanocyte dysfunction.

Leveraging epigenomes and three-dimensional genome organization for interpreting regulatory variation

The paper develops L-HiC-Reg to predict long-range regulatory interactions in new cell lines and applies it to interpret GWAS SNPs in complex diseases.

Chromatin structure and context-dependent sequence features control prime editing efficiency

This study identifies epigenomic and sequence features that influence prime editing efficiency and provides guidelines for optimizing prime editor designs.

Spectral clustering of single-cell multi-omics data on multilayer graphs

This paper introduces two spectral clustering algorithms for multi-omic single-cell data and shows they produce results comparable to Seurat's WNN method on a CITE-seq dataset.

Chd1 regulates repair of promoter-proximal DNA breaks to sustain hypertranscription in embryonic stem cells

This study identifies Chd1's role in repairing DNA breaks in embryonic stem cells, showing its importance in maintaining hypertranscription during development.

Predicting TCR-epitope Binding Specificity Using Deep Metric Learning and Multimodal Learning

This paper predicts TCR-epitope binding specificity using deep metric learning and multimodal learning, revealing key amino acid patterns and challenging assumptions about physical proximity.

Riemannian geometry and statistical modeling correct for batch effects and control false discoveries in single-cell surface protein count data from CITE-seq

Epigenetic engineering of yeast reveals dynamic molecular adaptation to methylation stress and genetic modulators of specific DNMT3 family members

This study uses yeast to investigate how DNA methylation is regulated by DNMT3 family members and identifies sequence preferences and spatial patterns of methylation.

SequencEnG: an Interactive Knowledge Base of Sequencing Techniques

SequencEnG provides an interactive knowledge base of 66 sequencing techniques and NGS analysis pipelines for educational purposes.

A unified computational framework for modeling genome-wide nucleosome landscape

This paper develops a computational framework to model genome-wide nucleosome landscape by quantifying sequence features' contributions to occupancy, translational, and rotational positioning in S. cerevisiae.

ClusterEnG: An interactive educational web resource for clustering big data

ClusterEnG provides an interactive web resource for clustering big data with visualizations and educational materials to help users understand clustering algorithms and their applications.

Maximum Entropy Methods for Extracting the Learned Features of Deep Neural Networks

This paper introduces a method to extract learned features from deep neural networks using maximum entropy principles, demonstrating its application in identifying transcription factor motifs and nucleosome positioning signals.

TeachEnG: a Teaching Engine for Genomics

TeachEnG is an interactive online tool that helps students learn sequence alignment and phylogenetic tree reconstruction through instructional games.

Emergent community agglomeration from data set geometry

This paper introduces effective dissimilarity transformation (EDT) to address the curse of dimensionality by transforming data geometry, improving clustering performance.