William Stafford Noble

The paper develops a deep learning-enhanced fragment ion indexing method to reduce memory usage in tandem mass spectrometry searches while maintaining search accuracy and speed.

The paper shows that structural variants from donor-specific assemblies influence chromatin contact divergence in humans, with many SVs linked to regulatory contexts and expression variation.

CasanovoGUI provides a user-friendly desktop interface for Casanovo, a deep learning model for de novo peptide sequencing, simplifying setup and result interpretation.

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.

The paper develops cell type-specific chromatin accessibility clocks using snATAC-seq data to predict age and detect aging-related changes in the brain.

scWeave translates between gene expression and chromatin structure at single-cell resolution using deep learning, improving cross-modality inference across multiple datasets.

HiCFoundation is a Hi-C-based model that integrates chromatin structure with epigenomic regulation across species, improving analysis of 3D genome architecture and single-cell data.

The paper introduces a label-free quantification feature in the Crux toolkit, demonstrating improved efficiency and accuracy compared to existing methods.

The paper develops a high-throughput Visual Cell Sorting workflow to link nuclear compartment morphology to single-nucleus transcriptomes, identifying a conserved transcriptional signature of nucleolar size linked to proteostasis.

The paper presents two large-scale proteomics datasets from mice exposed to varying X-ray doses and time points, enabling biomarker discovery and machine learning applications for radiation exposure assessment.

The paper introduces VIS-seq, an image-based method to assess variant effects on cellular phenotypes, revealing multidimensional variant impacts across different diseases.

MIMIR is a deep learning framework that imputes missing modalities and features in multi-omic data using shared representation learning, outperforming existing methods on pan-cancer data.

The paper investigates how sex chromosome aneuploidy affects gene expression and DNA methylation in different cell types, revealing cell-type specific dosage compensation mechanisms.

TwinC predicts trans-chromosomal DNA contacts using DNA sequence and provides insights into their functional mechanisms.

Carafe2 generates high-quality in silico spectral libraries for timsTOF DIA proteomics by fine-tuning deep learning models on DIA data, outperforming DDA-trained models in multiple applications.

The paper improves Casanovo, a deep learning de novo peptidesequencer, by enhancing interpretability, generalizing its use in database search, speeding up performance, and adding tools for adoption and result interpretation.

The paper integrates 4D Nucleome data to analyze 3D genome structure and function in human cells, revealing chromatin domains, loops, and interactions, and showing how genetic variants affect genome folding.

The paper introduces Lupine, a deep learning method for imputing missing values in TMT proteomics data, showing improved accuracy and identification of differentially expressed proteins across cancer datasets.

Puget predicts gene expression across cell types using sequence and Hi-C data, improving generalization to new biosamples and species.

Evo2HiC improves prediction of 3D genome architecture and epigenomic signals by integrating DNA sequences and Hi-C data, achieving better accuracy and computational efficiency.

Carafe generates high-quality in silico spectral libraries for DIA proteomics by training deep learning models directly on DIA data, improving fragment ion intensity prediction and peptide detection.

This study benchmarks mass spectrometry-based antibody sequencing methods, revealing biases related to sequence, concentration, protease, and bioinformatics tools.

HypoAD integrates MRI and snRNA-seq to analyze hypothalamic changes in aging and Alzheimer's disease, revealing volume and gene expression alterations in key subregions and cell types.

The paper predicts missing chromatin conformation contact maps across cell types and assays, exploring how 3D genome architecture varies.

The paper introduces an FDR control procedure for de novo peptide sequencing and demonstrates its validity and utility.

The paper improves the Tide search engine to handle large-scale mass spectrometry data efficiently, achieving faster speeds and lower memory usage than previous versions.

The paper introduces PIP-ECHO, a method to control false discovery rate in peptide-identity-propagation, showing it outperforms existing tools in FDR control and differential protein abundance detection.

This paper evaluates FDR control in mass spectrometry proteomics tools, finding that none consistently control FDR at the peptide level in DIA data, especially for single-cell datasets.

This paper introduces a method to improve quantitation in DIA proteomics by imputing peptide retention time boundaries, leading to more accurate quantifications and better estimation of radiation exposure.

The paper introduces Ledidi, a computational method for designing cis-regulatory elements by optimizing continuous sequences guided by deep learning models, validated experimentally for controlling transcriptional outcomes.