A Chemical-Genetic Interaction Matrix Reveals Drug Mechanism and Genetic Architecture

This dataset is associated with an article bearing the same name ( https://doi.org/10.64898/2026.01.28.701823 ), describing the results of >400 CRISPR/Cas9 chemical-genetic genome-wide knockout screens covering 305 different compounds and 5 conditions. The dataset also contains meta-data about each compound as well as the raw and/or processed data necessary to reproduce various analyses and figures from the article. The contents of each table and sub-table (tab) are described in the overview tab of each file. Table contents are briefly described below.
Table S1. Summary of the CGI dataset. (A) Sequenced samples. (B) Independent screens. (C) Compound details and verification. (D) NALM-6 gene expression. (E) CGIs detected.
Table S2. CRANKS scores.
Table S3. Known compound-target relationships.
Table S4. Combined CRANKS scores for two independent TRAIL screens.
Table S5. Jaccard similarity indices between screens in the CGI dataset. Pairs used in Fig. 2D are indicated by class.
Table S6. Compound-transporter CGI scores.
Table S7. (A) Compound dose-to-solubility ratios by screen. (B) Correlation coefficients of CRANKS scores with compound dose-to-solubility ratios.
Table S8. Vemurafenib SSL hits and Gene Ontology annotations.
Table S9. Predicted basic pKa and logP for compounds represented in the screen similarity network.
Table S10. Mean genome-wide CRANKS score profile of the 28 screens in the phospholipidosis cluster.
Table S11. LipidTOX Red assay determinations. (A) Cell counts and average LipidTOX levels. (B) Compound classification.
Table S12. Average CRANKS scores for combined screens. (A) Nocodazole. (B) Active site mTOR inhibitors. (C) Rapamycin. (D) AICAR. (E) Pyridostatin and topoisomerase II inhibitors. (F) DHFR inhibitor screens. (G) PARP inhibitors. (H) CHEK1 inhibitors.
Table S13. Comparison of CGIs for 9 shared compounds in this dataset and the Olivieri et al (2020) dataset.
Table S14. Data for assessment of compound synergy/antagonism, including luminescence readings, treatments by well and log2 Bliss scores determined by compound pair.
Table S15. (A) CRANKS2 scores. (B) Top 20,000 most correlated gene pairs.
Table S16. RNA-seq data. (A) FK-506 versus DMSO and read counts by gene. (B) PPP3CA knockout versus wild-type controls. (C) PPP3R1 knockout versus wild-type controls. (D) Transcription rates in PABIR1 knockout clones.
Table S17. List of DDR-associated genes.
Table S18. (A) Significant exon-level CGIs. (B) Exon classification and validation data.
Table S19. (A) Hypothetical gene CGIs with p-values <0.002 or absolute scores ≥2.5. (B) Hypothetical gene classification and mass spectral counts.
Table S20. Correlation between gene CRANKS scores and population doublings.