Vishaka Gopalan

STITCH pairs size-independent tangent-PCA contour shape modes with expression factors by canonical correlation, recovering keratinocyte, fibroblast and melanoma shape-expression links in Xenium and CosMx data.

The paper develops a mathematical model to study gene-expression dynamics and cell-state switching in melanoma using deterministic differential equations and cooperative binding approximations.

The paper develops CSI-TME to identify prognostic cell state interactions in IDH-mutant gliomas, revealing a pro-tumor network associated with immune checkpoint therapy response.

This study identifies early transcriptional signatures of drug resistance in cancer and across species, showing conservation and predictive value for treatment response.

IMMClock predicts immune age of T cells and NK cells using gene expression, showing immune aging is more linked to cellular aging than differentiation and is accelerated in diseases.

This study develops mouse models of brain metastatic melanoma to identify immune and molecular factors influencing response to immunotherapy.

This study identifies exosomal miRNAs that influence the pre-metastatic niche across multiple cancers, showing their association with survival and chemotherapy response.

This study applies a network-based method to identify key genes in BRCA subtypes and chemotherapy response in TNBC, revealing subtype-specific mediators and potential therapeutic targets.

The paper predicts gene expression changes after HDACi treatment using machine learning with pre-treatment transcriptome and epigenomic data, achieving high accuracy in distinguishing up- and downregulated genes.

This paper introduces Sgootr, a method to infer tumor lineage trees from single-cell methylation data, revealing metastatic progression models and identifying lineage-informative CpG sites.

miRSCAPE infers miRNA expression from RNA-seq data, showing high accuracy in bulk and single-cell datasets across multiple tissues and cell types.

The paper identifies a transcriptionally distinct subpopulation of healthy acinar cells with features of pancreatic progenitors and PDAC, linked to poor patient survival and age-related changes.

ChIPulate evaluates how biological and experimental factors affect ChIP-seq outcomes, finding that replicate numbers and extraction efficiency significantly impact binding strength inference.

The paper introduces CPI-EM, a method using ChIP-seq peak intensities and expectation maximization to detect cooperative TF pairs, showing improved performance over distance-based and sequence-based methods.