William E. Fondrie

Ptarmigan-1 uses contrastive learning to predict small molecule-protein interactions without requiring 3D structure, outperforming traditional methods on challenging targets.

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.

This paper introduces a method to profile the regulome across human cell lines and perturbations, revealing transcription factor dynamics and drug responses.

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 modifies the Casanovo model to account for digestion enzyme bias but finds no significant improvement in sequencing accuracy.

Casanovo translates tandem mass spectra into peptide sequences using a transformer model, outperforming existing methods and improving analysis in immunopeptidomics and metaproteomics.

The paper introduces FIRE to comprehensively map haplotype-selective chromatin accessibility in diploid human genomes, revealing somatic epimutations that contribute to non-genetic chromatin alterations.

Crema is an open-source Python tool that implements target-decoy competition methods for estimating false discovery rates in proteomics experiments.

This paper evaluates proteomics imputation methods using downstream-centric criteria and finds that MissForest performs best, though imputation may not improve differential expression detection.

The paper introduces nf-encyclopedia, a NextFlow pipeline for DIA proteomics analysis, showing it works well on cloud and local platforms and improves protein quantification with MSstats.

This paper integrates semi-supervised machine learning into the ANN-SoLo tool to improve spectral library searching for open modification peptides, increasing identification rates.

The paper summarizes recent developments in the Crux toolkit for analyzing bottom-up proteomics data, including speedups, new features, and tools.

Casanovo, a transformer encoder-decoder, translates unbinned tandem mass spectrum peaks into peptide sequences, beating DeepNovo and PointNovo on peptide-level precision across nine held-out species while running faster than both.

This paper introduces a workflow to build spectral libraries from DIA mass spectrometry data, showing it can detect phosphorylated peptides and outperforms library-free methods in accuracy and sensitivity.

The paper introduces ppx, a Python package for programmatic access to proteomics data repositories, demonstrating its utility in reproducible research workflows.

mokapot improves peptide detection in proteomics by offering a flexible semi-supervised learning approach, enhancing accuracy in RNA-binding protein and single-cell studies.

The paper proposes a static modeling approach for Percolator to improve peptide detection in small-scale proteomics experiments by leveraging large datasets.