Effects of threshold choice on biological conclusions reached during analysis of gene expression by DNA microarrays
- PMID: 15951424
- PMCID: PMC1149502
- DOI: 10.1073/pnas.0502674102
Effects of threshold choice on biological conclusions reached during analysis of gene expression by DNA microarrays
Abstract
Global analysis of gene expression by using DNA microarrays is employed increasingly to search for differences in biological properties between normal and diseased tissue. In such studies, expression that deviates from defined thresholds commonly is used for creating genetic signatures that characterize disease vs. normality. Although it is axiomatic that the threshold parameters applied to microarray analysis will alter the contents of such genetic signatures, the extent to which threshold choice can affect the fundamental conclusions made from microarray-based studies has not been elucidated. We used GABRIEL (Genetic Analysis By Rules Incorporating Expert Logic), a platform of knowledge-based algorithms for the global analysis of gene expression, together with conventional statistical approaches, to examine the sensitivity of conclusions to threshold choice in recently published microarray-based studies. An analysis of the effects of threshold decisions in one of these studies [Ramaswamy, S., Ross, K. N., Lander, E. S. & Golub, T. R. (2003) Nat. Genet. 33, 49-54], which arrived at the important conclusion that the metastatic potential of primary tumors is encoded by the bulk of cells in the tumor, is the focus of this article. We discovered that support for this conclusion highly depends on the threshold used to create gene expression signatures. We also found that threshold choice dramatically affected the gene function categories represented nonrandomly in signatures. Our results suggest that the robustness of biological conclusions made by using microarray analysis should be routinely assessed by examining the validity of the conclusions by using a range of threshold parameters.
Figures
Similar articles
-
Large scale real-time PCR validation on gene expression measurements from two commercial long-oligonucleotide microarrays.BMC Genomics. 2006 Mar 21;7:59. doi: 10.1186/1471-2164-7-59. BMC Genomics. 2006. PMID: 16551369 Free PMC article.
-
Independent component analysis-based penalized discriminant method for tumor classification using gene expression data.Bioinformatics. 2006 Aug 1;22(15):1855-62. doi: 10.1093/bioinformatics/btl190. Epub 2006 May 18. Bioinformatics. 2006. PMID: 16709589
-
Gene expression profiling reveals reproducible human lung adenocarcinoma subtypes in multiple independent patient cohorts.J Clin Oncol. 2006 Nov 1;24(31):5079-90. doi: 10.1200/JCO.2005.05.1748. J Clin Oncol. 2006. PMID: 17075127 Review.
-
Using microarray analysis as a prognostic and predictive tool in oncology: focus on breast cancer and normal tissue toxicity.Semin Radiat Oncol. 2008 Apr;18(2):105-14. doi: 10.1016/j.semradonc.2007.10.007. Semin Radiat Oncol. 2008. PMID: 18314065 Review.
-
Analysis of DNA microarrays using algorithms that employ rule-based expert knowledge.Proc Natl Acad Sci U S A. 2002 Feb 19;99(4):2118-23. doi: 10.1073/pnas.251687398. Proc Natl Acad Sci U S A. 2002. PMID: 11854507 Free PMC article.
Cited by
-
A novel pathway analysis approach based on the unexplained disregulation of genes.Proc IEEE Inst Electr Electron Eng. 2017 Mar;105(3):482-495. doi: 10.1109/JPROC.2016.2531000. Epub 2016 Mar 24. Proc IEEE Inst Electr Electron Eng. 2017. PMID: 30337764 Free PMC article.
-
A Powerful Test for SNP Effects on Multivariate Binary Outcomes using Kernel Machine Regression.Stat Biosci. 2018 Apr;10(1):117-138. doi: 10.1007/s12561-017-9189-9. Epub 2017 Mar 24. Stat Biosci. 2018. PMID: 30420901 Free PMC article.
-
Gene set analysis exploiting the topology of a pathway.BMC Syst Biol. 2010 Sep 1;4:121. doi: 10.1186/1752-0509-4-121. BMC Syst Biol. 2010. PMID: 20809931 Free PMC article.
-
Assessing the biological significance of gene expression signatures and co-expression modules by studying their network properties.PLoS One. 2011 Mar 7;6(3):e17474. doi: 10.1371/journal.pone.0017474. PLoS One. 2011. PMID: 21408226 Free PMC article.
-
Network-based analysis of affected biological processes in type 2 diabetes models.PLoS Genet. 2007 Jun;3(6):e96. doi: 10.1371/journal.pgen.0030096. PLoS Genet. 2007. PMID: 17571924 Free PMC article.
References
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Other Literature Sources