Shreyash Sonthalia

Bromo, a pairwise transformer, ranks a protein's peptide precursors by expected mass spectrometry response, beating XGBoost, PREGO and PeptideRanger on held-out yeast data, and can be fine-tuned per instrument.

CellCover identifies marker gene panels for cell types by solving a set-covering problem, outperforming differential expression methods in capturing cell-specific signals and developmental progression.

The paper compiles a large transcriptomic dataset of neocortical development and uses joint matrix decomposition to identify conserved transcriptome dynamics across mammals and developmental stages.

AtomWorks, a modular atom-level data framework, trains structure prediction and design models; the resulting RF3 improves chirality handling and performs between AF3 and Boltz.

The paper analyzes single-cell transcriptomics data of iPSC-derived neurons and astrocytes to assess their similarity to in vivo counterparts and their disease-related gene expression patterns.

SJD identifies shared molecular variation across multi-omics datasets by decomposing matrices with hierarchical structure, applied to neurogenesis RNA-seq data.

The paper describes the Neuroscience Multi-Omic Archive, a repository for single-cell transcriptomic and epigenomic data from mammalian brains, providing tools for data access and analysis.