Recent therapeutic successes have renewed interest in drug combinations, but experimental screening approaches are costly and often identify only small numbers of synergistic combinations. The DREAM consortium launched an open challenge to foster the development of in silico methods to computationally rank 91 compound pairs, from the most synergistic to the most antagonistic, based on gene-expression profiles of human B cells treated with individual compounds at multiple time points and concentrations. Using scoring metrics based on experimental dose-response curves, we assessed 32 methods (31 community-generated approaches and SynGen), four of which performed significantly better than random guessing. We highlight similarities between the methods. Although the accuracy of predictions was not optimal, we find that computational prediction of compound-pair activity is possible, and that community challenges can be useful to advance the field of in silico compound-synergy prediction.
%0 Journal Article
%1 Bansal2014Community
%A Bansal, Mukesh
%A Yang, Jichen
%A Karan, Charles
%A Menden, Michael P.
%A Costello, James C.
%A Tang, Hao
%A Xiao, Guanghua
%A Li, Yajuan
%A Allen, Jeffrey
%A Zhong, Rui
%A Chen, Beibei
%A Kim, Minsoo
%A Wang, Tao
%A Heiser, Laura M.
%A Realubit, Ronald
%A Mattioli, Michela
%A Alvarez, Mariano J.
%A Shen, Yao
%A NCI-DREAM Community,
%A Gallahan, Daniel
%A Singer, Dinah
%A Saez-Rodriguez, Julio
%A Xie, Yang
%A Stolovitzky, Gustavo
%A Califano, Andrea
%A NCI-DREAM Community,
%D 2014
%J Nature biotechnology
%K dream drug-synergy
%N 12
%P 1213--1222
%T A community computational challenge to predict the activity of pairs of compounds.
%U http://view.ncbi.nlm.nih.gov/pubmed/25419740
%V 32
%X Recent therapeutic successes have renewed interest in drug combinations, but experimental screening approaches are costly and often identify only small numbers of synergistic combinations. The DREAM consortium launched an open challenge to foster the development of in silico methods to computationally rank 91 compound pairs, from the most synergistic to the most antagonistic, based on gene-expression profiles of human B cells treated with individual compounds at multiple time points and concentrations. Using scoring metrics based on experimental dose-response curves, we assessed 32 methods (31 community-generated approaches and SynGen), four of which performed significantly better than random guessing. We highlight similarities between the methods. Although the accuracy of predictions was not optimal, we find that computational prediction of compound-pair activity is possible, and that community challenges can be useful to advance the field of in silico compound-synergy prediction.
@article{Bansal2014Community,
abstract = {Recent therapeutic successes have renewed interest in drug combinations, but experimental screening approaches are costly and often identify only small numbers of synergistic combinations. The {DREAM} consortium launched an open challenge to foster the development of in silico methods to computationally rank 91 compound pairs, from the most synergistic to the most antagonistic, based on gene-expression profiles of human B cells treated with individual compounds at multiple time points and concentrations. Using scoring metrics based on experimental dose-response curves, we assessed 32 methods (31 community-generated approaches and {SynGen}), four of which performed significantly better than random guessing. We highlight similarities between the methods. Although the accuracy of predictions was not optimal, we find that computational prediction of compound-pair activity is possible, and that community challenges can be useful to advance the field of in silico compound-synergy prediction.},
added-at = {2018-12-02T16:09:07.000+0100},
author = {Bansal, Mukesh and Yang, Jichen and Karan, Charles and Menden, Michael P. and Costello, James C. and Tang, Hao and Xiao, Guanghua and Li, Yajuan and Allen, Jeffrey and Zhong, Rui and Chen, Beibei and Kim, Minsoo and Wang, Tao and Heiser, Laura M. and Realubit, Ronald and Mattioli, Michela and Alvarez, Mariano J. and Shen, Yao and {NCI-DREAM Community} and Gallahan, Daniel and Singer, Dinah and Saez-Rodriguez, Julio and Xie, Yang and Stolovitzky, Gustavo and Califano, Andrea and {NCI-DREAM Community}},
biburl = {https://www.bibsonomy.org/bibtex/2e444e3a241c975104b29bf0739b3aca1/karthikraman},
citeulike-article-id = {14387651},
citeulike-linkout-0 = {http://view.ncbi.nlm.nih.gov/pubmed/25419740},
citeulike-linkout-1 = {http://www.hubmed.org/display.cgi?uids=25419740},
interhash = {ba85f3a753f37066493054570ae89cfa},
intrahash = {e444e3a241c975104b29bf0739b3aca1},
issn = {1546-1696},
journal = {Nature biotechnology},
keywords = {dream drug-synergy},
month = dec,
number = 12,
pages = {1213--1222},
pmid = {25419740},
posted-at = {2017-07-04 05:36:07},
priority = {2},
timestamp = {2018-12-02T16:09:07.000+0100},
title = {A community computational challenge to predict the activity of pairs of compounds.},
url = {http://view.ncbi.nlm.nih.gov/pubmed/25419740},
volume = 32,
year = 2014
}