A community effort to assess and improve drug sensitivity prediction algorithms

NCI-DREAM Community, Hao Tang

Research output: Contribution to journalArticle

227 Citations (Scopus)

Abstract

Predicting the best treatment strategy from genomic information is a core goal of precision medicine. Here we focus on predicting drug response based on a cohort of genomic, epigenomic and proteomic profiling data sets measured in human breast cancer cell lines. Through a collaborative effort between the National Cancer Institute (NCI) and the Dialogue on Reverse Engineering Assessment and Methods (DREAM) project, we analyzed a total of 44 drug sensitivity prediction algorithms. The top-performing approaches modeled nonlinear relationships and incorporated biological pathway information. We found that gene expression microarrays consistently provided the best predictive power of the individual profiling data sets; however, performance was increased by including multiple, independent data sets. We discuss the innovations underlying the top-performing methodology, Bayesian multitask MKL, and we provide detailed descriptions of all methods. This study establishes benchmarks for drug sensitivity prediction and identifies approaches that can be leveraged for the development of new methods.

Original languageEnglish (US)
Pages (from-to)1202-1212
Number of pages11
JournalNature Biotechnology
Volume32
Issue number12
DOIs
StatePublished - Dec 1 2014

Fingerprint

Reverse engineering
Microarrays
Gene expression
Pharmaceutical Preparations
Medicine
Benchmarking
Precision Medicine
National Cancer Institute (U.S.)
Innovation
Cells
Epigenomics
Proteomics
Breast Neoplasms
Gene Expression
Cell Line
Datasets
Therapeutics

ASJC Scopus subject areas

  • Applied Microbiology and Biotechnology
  • Biotechnology
  • Molecular Medicine
  • Bioengineering
  • Biomedical Engineering

Cite this

A community effort to assess and improve drug sensitivity prediction algorithms. / NCI-DREAM Community; Tang, Hao.

In: Nature Biotechnology, Vol. 32, No. 12, 01.12.2014, p. 1202-1212.

Research output: Contribution to journalArticle

NCI-DREAM Community ; Tang, Hao. / A community effort to assess and improve drug sensitivity prediction algorithms. In: Nature Biotechnology. 2014 ; Vol. 32, No. 12. pp. 1202-1212.
@article{c4d04229761841caa376c8526b0bf165,
title = "A community effort to assess and improve drug sensitivity prediction algorithms",
abstract = "Predicting the best treatment strategy from genomic information is a core goal of precision medicine. Here we focus on predicting drug response based on a cohort of genomic, epigenomic and proteomic profiling data sets measured in human breast cancer cell lines. Through a collaborative effort between the National Cancer Institute (NCI) and the Dialogue on Reverse Engineering Assessment and Methods (DREAM) project, we analyzed a total of 44 drug sensitivity prediction algorithms. The top-performing approaches modeled nonlinear relationships and incorporated biological pathway information. We found that gene expression microarrays consistently provided the best predictive power of the individual profiling data sets; however, performance was increased by including multiple, independent data sets. We discuss the innovations underlying the top-performing methodology, Bayesian multitask MKL, and we provide detailed descriptions of all methods. This study establishes benchmarks for drug sensitivity prediction and identifies approaches that can be leveraged for the development of new methods.",
author = "{NCI-DREAM Community} and Costello, {James C.} and Heiser, {Laura M.} and Elisabeth Georgii and Mehmet G{\"o}nen and Menden, {Michael P.} and Wang, {Nicholas J.} and Mukesh Bansal and Muhammad Ammad-Ud-Din and Petteri Hintsanen and Khan, {Suleiman A.} and Mpindi, {John Patrick} and Olli Kallioniemi and Antti Honkela and Tero Aittokallio and Krister Wennerberg and Collins, {James J.} and Dan Gallahan and Dinah Singer and Julio Saez-Rodriguez and Samuel Kaski and Gray, {Joe W.} and Gustavo Stolovitzky and Abbuehl, {Jean Paul} and Jeffrey Allen and Altman, {Russ B.} and Shawn Balcome and Alexis Battle and Andreas Bender and Bonnie Berger and Jonathan Bernard and Madhuchhanda Bhattacharjee and Krithika Bhuvaneshwar and Bieberich, {Andrew A.} and Fred Boehm and Andrea Califano and Christina Chan and Beibei Chen and Chen, {Ting Huei} and Jaejoon Choi and Coelho, {Luis Pedro} and Thomas Cokelaer and Collins, {James C.} and Creighton, {Chad J.} and Jike Cui and Will Dampier and Davisson, {V. Jo} and Hao Tang and Raamesh Deshpande and Guanghua Xiao and Yang Xie",
year = "2014",
month = "12",
day = "1",
doi = "10.1038/nbt.2877",
language = "English (US)",
volume = "32",
pages = "1202--1212",
journal = "Biotechnology",
issn = "0733-222X",
publisher = "Nature Publishing Group",
number = "12",

}

TY - JOUR

T1 - A community effort to assess and improve drug sensitivity prediction algorithms

AU - NCI-DREAM Community

AU - Costello, James C.

AU - Heiser, Laura M.

AU - Georgii, Elisabeth

AU - Gönen, Mehmet

AU - Menden, Michael P.

AU - Wang, Nicholas J.

AU - Bansal, Mukesh

AU - Ammad-Ud-Din, Muhammad

AU - Hintsanen, Petteri

AU - Khan, Suleiman A.

AU - Mpindi, John Patrick

AU - Kallioniemi, Olli

AU - Honkela, Antti

AU - Aittokallio, Tero

AU - Wennerberg, Krister

AU - Collins, James J.

AU - Gallahan, Dan

AU - Singer, Dinah

AU - Saez-Rodriguez, Julio

AU - Kaski, Samuel

AU - Gray, Joe W.

AU - Stolovitzky, Gustavo

AU - Abbuehl, Jean Paul

AU - Allen, Jeffrey

AU - Altman, Russ B.

AU - Balcome, Shawn

AU - Battle, Alexis

AU - Bender, Andreas

AU - Berger, Bonnie

AU - Bernard, Jonathan

AU - Bhattacharjee, Madhuchhanda

AU - Bhuvaneshwar, Krithika

AU - Bieberich, Andrew A.

AU - Boehm, Fred

AU - Califano, Andrea

AU - Chan, Christina

AU - Chen, Beibei

AU - Chen, Ting Huei

AU - Choi, Jaejoon

AU - Coelho, Luis Pedro

AU - Cokelaer, Thomas

AU - Collins, James C.

AU - Creighton, Chad J.

AU - Cui, Jike

AU - Dampier, Will

AU - Davisson, V. Jo

AU - Tang, Hao

AU - Deshpande, Raamesh

AU - Xiao, Guanghua

AU - Xie, Yang

PY - 2014/12/1

Y1 - 2014/12/1

N2 - Predicting the best treatment strategy from genomic information is a core goal of precision medicine. Here we focus on predicting drug response based on a cohort of genomic, epigenomic and proteomic profiling data sets measured in human breast cancer cell lines. Through a collaborative effort between the National Cancer Institute (NCI) and the Dialogue on Reverse Engineering Assessment and Methods (DREAM) project, we analyzed a total of 44 drug sensitivity prediction algorithms. The top-performing approaches modeled nonlinear relationships and incorporated biological pathway information. We found that gene expression microarrays consistently provided the best predictive power of the individual profiling data sets; however, performance was increased by including multiple, independent data sets. We discuss the innovations underlying the top-performing methodology, Bayesian multitask MKL, and we provide detailed descriptions of all methods. This study establishes benchmarks for drug sensitivity prediction and identifies approaches that can be leveraged for the development of new methods.

AB - Predicting the best treatment strategy from genomic information is a core goal of precision medicine. Here we focus on predicting drug response based on a cohort of genomic, epigenomic and proteomic profiling data sets measured in human breast cancer cell lines. Through a collaborative effort between the National Cancer Institute (NCI) and the Dialogue on Reverse Engineering Assessment and Methods (DREAM) project, we analyzed a total of 44 drug sensitivity prediction algorithms. The top-performing approaches modeled nonlinear relationships and incorporated biological pathway information. We found that gene expression microarrays consistently provided the best predictive power of the individual profiling data sets; however, performance was increased by including multiple, independent data sets. We discuss the innovations underlying the top-performing methodology, Bayesian multitask MKL, and we provide detailed descriptions of all methods. This study establishes benchmarks for drug sensitivity prediction and identifies approaches that can be leveraged for the development of new methods.

UR - http://www.scopus.com/inward/record.url?scp=84906549588&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84906549588&partnerID=8YFLogxK

U2 - 10.1038/nbt.2877

DO - 10.1038/nbt.2877

M3 - Article

VL - 32

SP - 1202

EP - 1212

JO - Biotechnology

JF - Biotechnology

SN - 0733-222X

IS - 12

ER -