Determinants of cysteine coverage by chemoproteomics
- 1Crick-GSK Biomedical LinkLabs, GSK
Abstract
Covalent chemoproteomics has become a powerful route to expanding the druggable proteome, enabling proteome-wide screening of compounds that covalently modify nucleophilic residues, most commonly cysteine. Yet typical experiments detect only 5-18% of the ~250,000 cysteines encoded by the human proteome, and the determinants of this limited coverage remain poorlydefined. Here we combine cysteine featurisation, explainable machine learning (ML) and robust statistical filtering to establish what governs cysteine detectability in iodoacetamide- desthiobiotin (IA-DTB) chemoproteomics. Using a scalable feature-generation pipeline, we computed physicochemical and structural properties for cysteines and the corresponding tryptic peptides and trained models that predict detectability with high accuracy. Shapley analysis identifies protein abundance and intrinsic peptide properties governing LC-MS behaviour as the dominant determinants, whereas cysteine structural descriptors contribute comparatively little, indicating that detection is constrained downstream of probe reactivity. We further introduce a distance-based statistical filter that improves reproducibility across replicates, and show that cell line selection, subcellular fractionation and alternative electrophiles together expand coverage by >30%. More broadly, this work moves optimisation of chemoproteomics workflows from empirical trial-and-error towards rational, data-driven design.
Related
- ERK-mediated phosphorylation modulates the autolytic processing of the cysteine protease CAPN3
- An atypical cysteine-type protein domain regulates cuticular extracellular matrix assembly in nematodes
- CYSTM1, a drought-responsive cysteine-rich transmembrane module protein, negatively affects drought tolerance in Arabidopsis thaliana
- Bronchial epithelial IB3-1 cells exposed to the anti-SARS-CoV2 BNT162b2 vaccine: induction of apoptosis is partially reversed by Aged Garlic Extract (AGE) and its bioactive components S-allyl-Cysteine and S1-propenyl-l-Cysteine
- Crystal Structure of Bacillus subtilis Cysteine Desulfurase SufS and Its Dynamic Interaction with Frataxin and Scaffold Protein SufU.
BibTeX
@misc{riziotis2026determinants,
title = {{Determinants of cysteine coverage by chemoproteomics}},
author = {I. G. Riziotis and H. Wilders and A. Cairoli and H. Lambert and J. Dark and G. S. Biggs and E. E. Cawood and W. J. McCarthy and L. Nightingale and P. Magarinos and M. Gavrielides and K. Ambrose and A. Strange and M. J. Skehel and S. J. Boulton and J. Lumley and D. House and K. Rittinger and J. T. Bush},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.12.758458},
url = {https://www.biorxiv.org/content/10.64898/2026.10.12.758458v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Determinants of cysteine coverage by chemoproteomics
AU - I. G. Riziotis
AU - H. Wilders
AU - A. Cairoli
AU - H. Lambert
AU - J. Dark
AU - G. S. Biggs
AU - E. E. Cawood
AU - W. J. McCarthy
AU - L. Nightingale
AU - P. Magarinos
AU - M. Gavrielides
AU - K. Ambrose
AU - A. Strange
AU - M. J. Skehel
AU - S. J. Boulton
AU - J. Lumley
AU - D. House
AU - K. Rittinger
AU - J. T. Bush
PY - 2026
DA - 2026/10/12
PB - bioRxiv
DO - 10.64898/2026.10.12.758458
UR - https://www.biorxiv.org/content/10.64898/2026.10.12.758458v1
AB - Covalent chemoproteomics has become a powerful route to expanding the druggable proteome, enabling proteome-wide screening of compounds that covalently modify nucleophilic residues, most commonly cysteine. Yet typical experiments detect only 5-18% of the ~250,000 cysteines encoded by the human proteome, and the determinants of this limited coverage remain poorlydefined. Here we combine cysteine featurisation, explainable machine learning (ML) and robust statistical filtering to establish what governs cysteine detectability in iodoacetamide- desthiobiotin (IA-DTB) chemoproteomics. Using a scalable feature-generation pipeline, we computed physicochemical and structural properties for cysteines and the corresponding tryptic peptides and trained models that predict detectability with high accuracy. Shapley analysis identifies protein abundance and intrinsic peptide properties governing LC-MS behaviour as the dominant determinants, whereas cysteine structural descriptors contribute comparatively little, indicating that detection is constrained downstream of probe reactivity. We further introduce a distance-based statistical filter that improves reproducibility across replicates, and show that cell line selection, subcellular fractionation and alternative electrophiles together expand coverage by >30%. More broadly, this work moves optimisation of chemoproteomics workflows from empirical trial-and-error towards rational, data-driven design.
ER -