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 Jun 21;10:68. doi: 10.1038/s41540-024-00394-w

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

© The Author(s) 2024, corrected publication 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. 6 — A In silico balanced accuracy across various synergy thresholds. Matrix represents the threshold value-dependent consensus (expressed in Balanced Accuracy, BA= (TPR + TNR) /2) between in vitro and in silico synergy. In the DREAM challenge, a threshold of 20 for both in vitro and in silico synergy was used to distinguish synergistic and non-synergistic combinations (Methods). However, the DREAM in vitro results were scaled between 0 and 100, while our in silico predictions are between 0 and 1, therefore the in silico threshold of 0.2 would correspond to it in this case. If we would use this cutoff value for the same endeavor as a gold standard, the balanced accuracy would be just under 0.6 (BA = 0.575, sensitivity=0.37, specificity=0.78). When we set the in vitro threshold for synergy to 30 and 20 for in silico synergy, our predictive power in balanced accuracy significantly improved (BA = 0.62, sensitivity=0.46, specificity=0.78). By increasing the in vitro threshold, the number of synergistic data points decreased, and the model was able to correctly detect a bigger proportion of synergistic combinations. This shows that in silico predictions are more accurate when stronger synergy is observed in vitro. B Proportion of accurate and inaccurate simulated measurements in different combination MoA groups. Due to the low in vitro data coverage, in many cases, benchmarking was not possible. Both in vitro and in silico synergy thresholds were set at 20, e.g., if both values were bigger or equal to 20, we considered that combination-cell line pair to be accurately synergistic and vice versa. Proportion of accurate and inaccurate simulations was calculated compared to the 977 overall data points. MoA groups are sorted based on the descending proportion of accurate simulations. CSNK Casein kinase, DDR DNA damage repair, EM/TM epigenetic/transcriptomic modulation, ERK extracellular signal-regulated kinase, JAK/STAT Janus kinase/signal transducers and activators of transcription, NFKB NF-kappa B, PIK3/AKT/MTOR phosphatidylinositol 3’ -kinase(PI3K)-AKT-mTOR, TK/RTK tyrosine kinase/receptor tyrosine kinase.