Lack of sufficiently strong informative features limits the potential of gene expression analysis as predictive tool for many clinical classification problems
-
Published:2011-12
Issue:1
Volume:12
Page:
-
ISSN:1471-2105
-
Container-title:BMC Bioinformatics
-
language:en
-
Short-container-title:BMC Bioinformatics
Author:
Hess Kenneth R,Wei Caimiao,Qi Yuan,Iwamoto Takayuki,Symmans W Fraser,Pusztai Lajos
Abstract
Abstract
Background
Our goal was to examine how various aspects of a gene signature influence the success of developing multi-gene prediction models. We inserted gene signatures into three real data sets by altering the expression level of existing probe sets. We varied the number of probe sets perturbed (signature size), the fold increase of mean probe set expression in perturbed compared to unperturbed data (signature strength) and the number of samples perturbed. Prediction models were trained to identify which cases had been perturbed. Performance was estimated using Monte-Carlo cross validation.
Results
Signature strength had the greatest influence on predictor performance. It was possible to develop almost perfect predictors with as few as 10 features if the fold difference in mean expression values were > 2 even when the spiked samples represented 10% of all samples. We also assessed the gene signature set size and strength for 9 real clinical prediction problems in six different breast cancer data sets.
Conclusions
We found sufficiently large and strong predictive signatures only for distinguishing ER-positive from ER-negative cancers, there were no strong signatures for more subtle prediction problems. Current statistical methods efficiently identify highly informative features in gene expression data if such features exist and accurate models can be built with as few as 10 highly informative features. Features can be considered highly informative if at least 2-fold expression difference exists between comparison groups but such features do not appear to be common for many clinically relevant prediction problems in human data sets.
Publisher
Springer Science and Business Media LLC
Subject
Applied Mathematics,Computer Science Applications,Molecular Biology,Biochemistry,Structural Biology
Reference30 articles.
1. van de Vijver MJ, He YD, van't Veer LJ, Dai H, Hart AA, Voskuil DW, Schreiber GJ, Peterse JL, Roberts C, Marton MJ, et al.: A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 2002, 347(25):1999–2009. 10.1056/NEJMoa021967 2. Wang Y, Klijn JG, Zhang Y, Sieuwerts AM, Look MP, Yang F, Talantov D, Timmermans M, Meijer-van Gelder ME, Yu J, et al.: Gene-expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer. Lancet 2005, 365(9460):671–679. 3. Buyse M, Loi S, van't Veer L, Viale G, Delorenzi M, Glas AM, d'Assignies MS, Bergh J, Lidereau R, Ellis P, et al.: Validation and clinical utility of a 70-gene prognostic signature for women with node-negative breast cancer. J Natl Cancer Inst 2006, 98(17):1183–1192. 10.1093/jnci/djj329 4. Hess KR, Anderson K, Symmans WF, Valero V, Ibrahim N, Mejia JA, Booser D, Theriault RL, Buzdar AU, Dempsey PJ, et al.: Pharmacogenomic predictor of sensitivity to preoperative chemotherapy with paclitaxel and fluorouracil, doxorubicin, and cyclophosphamide in breast cancer. J Clin Oncol 2006, 24(26):4236–4244. 10.1200/JCO.2006.05.6861 5. Cleator S, Tsimelzon A, Ashworth A, Dowsett M, Dexter T, Powles T, Hilsenbeck S, Wong H, Osborne CK, O'Connell P, et al.: Gene expression patterns for doxorubicin (Adriamycin) and cyclophosphamide (cytoxan) (AC) response and resistance. Breast Cancer Res Treat 2006, 95(3):229–233. 10.1007/s10549-005-9009-7
Cited by
11 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
|
|