Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2024 Sep 15;7:1149. doi: 10.1038/s42003-024-06865-4

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2024

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

PMC Copyright notice

Fig. 5 — a Distinct protein target clusters emerge from UMAP decomposition of adjusted attribution vectors at compound IC50s for predicted and fitted dose-response relationships in MDAMB231, MDAMB231-LM2, HCC1806, HCC1806-LM2b/c, SW480, and SW480-LvM2 prospective cell lines. Control compound set (CCS) attribution vectors colored by nominal target class. b Comparison of adjusted mutual information (AMI) derived from CCS nominal target labels and K-means clustering of trained model adjusted attribution vectors (n = 5 independent samples), compound fingerprints (n = 5 independent samples), random-model adjusted attribution vectors (n = 5 independent samples) and permuted-model adjusted attribution vectors (n = 5 independent samples). Centerline, median; box limits, upper and lower quartiles; whiskers, 1.5x interquartile range; points, outliers; two-sided Wilcoxon rank-sum test; *p < =5e-2, **p < =1e-2. c Average attribution difference between the highest significance target of the nominal target class versus all other target classes. Two-sided Wilcoxon rank-sum test; *p < =5e-2, **p < =1e-2, ***p < =1e-3, ****p < =1e-4. Sample sizes: n = 14 independent case samples, n = 455 independent control samples except otherwise noted; EGFR;ERBB2 (n = 21 case, n = 448 control); HDAC1/2/3/6/8 (n = 56 case, n = 413 control); HMGCR (n = 21 case, n = 448 control); NAMPT (n = 28 case, n = 441 control); TP53 (n = 21 case, n = 448 control). d Leiden clustering of all attribution vectors. e Comparison of PPI subgraph connectivity derived from clustered target profiles (n = 26 independent samples), random target profiles (n = 26 independent samples), and random protein-coding genes (n = 26 independent samples). Centerline, median; box limits, upper and lower quartiles; whiskers, 1.5x interquartile range; points, outliers; two-sided Wilcoxon rank-sum test; ***p < =1e-3, ****p < =1e-4. f Network representation of exemplar clustered target profile subgraphs.