Gene essentiality signatures for exploring cancer-relevant targets and mechanisms in large-scale drug sensitivity screens

Cancer drugs can produce phenotypic effects beyond their annotated targets, complicating mechanism-of-action (MoA) inference. We introduce gene-essentiality-derived signatures as a data-centric representation of drugs, learned by regressing cell-line drug sensitivity against genome-wide loss-of-function dependency profiles across hundreds of cancer cell lines. These interpretable vectors capture associations between drug response and gene dependencies. In supervised target prediction, the signatures outperformed chemical fingerprints and drug-perturbation transcriptomic signatures. Without target priors, top-ranked genes and their first-degree protein-protein interaction neighbors were enriched for known targets and pathway components, enabling de novo MoA exploration, including EGFR/HER2 signaling for lapatinib and DNA mismatch repair-linked compounds. We applied the framework to PRISM cancer cell-line screening data to predict cancer-relevant targets and MoAs for non-oncology drugs. Multiple predictions received orthogonal support from proteome-wide target deconvolution and structure-based docking, highlighting opportunities for oncology-focused drug repurposing. These interpretable, scalable embeddings support target and MoA hypothesis generation.

Authors

Institutions

Publication Details

Journal
iScience
Published
2026-11-21
DOI
https://doi.org/10.1016/j.isci.2026.118287
Primary Topic
Computational Drug Discovery Methods
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Gene essentiality signatures for exploring cancer-relevant targets and mechanisms in large-scale drug sensitivity screens

Jie Bao, Jehad Aldahdooh, Yinyin Wang, Ziaurrehman Tanoli et al.
iScience
Computational Drug Discovery Methods
article

Gene essentiality signatures for exploring cancer-relevant targets and mechanisms in large-scale drug sensitivity screens

Jie Bao, Jehad Aldahdooh, Yinyin Wang, Ziaurrehman Tanoli, Chuqi Lei, Wenyu Wang, Xuepei Zhang, Johanna Eriksson, Shan Huang, Massimiliano Gaetani, Min Li, Jing Tang, Shuyu Zheng
article en

Abstract

Cancer drugs can produce phenotypic effects beyond their annotated targets, complicating mechanism-of-action (MoA) inference. We introduce gene-essentiality-derived signatures as a data-centric representation of drugs, learned by regressing cell-line drug sensitivity against genome-wide loss-of-function dependency profiles across hundreds of cancer cell lines. These interpretable vectors capture associations between drug response and gene dependencies. In supervised target prediction, the signatures outperformed chemical fingerprints and drug-perturbation transcriptomic signatures. Without target priors, top-ranked genes and their first-degree protein-protein interaction neighbors were enriched for known targets and pathway components, enabling de novo MoA exploration, including EGFR/HER2 signaling for lapatinib and DNA mismatch repair-linked compounds. We applied the framework to PRISM cancer cell-line screening data to predict cancer-relevant targets and MoAs for non-oncology drugs. Multiple predictions received orthogonal support from proteome-wide target deconvolution and structure-based docking, highlighting opportunities for oncology-focused drug repurposing. These interpretable, scalable embeddings support target and MoA hypothesis generation.

iScienceVol. 29(12)
University of Helsinki (FI), Central South University (CN), Science for Life Laboratory (SE), Karolinska Institutet (SE), Second Affiliated Hospital of Xi'an Jiaotong University (CN)
Computational Drug Discovery Methods
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon