An Integrated Approach for the Analysis of Biological Pathways using Mixed Models

• Published in: PLoS genetics Vol. 4; no. 7; p. e1000115
• Main Authors: Wang, Lily, Zhang, Bing, Wolfinger, Russell D., Chen, Xi
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 01.07.2008; Public Library of Science (PLoS)
• Subjects: Aldehydes - administration & dosage; Area Under Curve; Arrays; Cell Line, Tumor; Computational Biology/Metabolic Networks; Computational Biology/Molecular Genetics; Computational Biology/Signaling Networks; Databases, Genetic; Design; Diabetes Mellitus - genetics; Dose-Response Relationship, Drug; Experiments; Gene Expression - drug effects; Gene Expression Profiling - statistics & numerical data; Genomics; Humans; Linear Models; Methods; Models, Genetic; Models, Statistical; Molecular Biology/Bioinformatics; Oligonucleotide Array Sequence Analysis - statistics & numerical data; Ontology; Sensitivity and Specificity; Studies; Systems Biology
• ISSN: 1553-7404; 1553-7390; 1553-7404
• DOI: 10.1371/journal.pgen.1000115

Gene class, ontology, or pathway testing analysis has become increasingly popular in microarray data analysis. Such approaches allow the integration of gene annotation databases, such as Gene Ontology and KEGG Pathway, to formally test for subtle but coordinated changes at a system level. Higher power in gene class testing is gained by combining weak signals from a number of individual genes in each pathway. We propose an alternative approach for gene-class testing based on mixed models, a class of statistical models that: a) provides the ability to model and borrow strength across genes that are both up and down in a pathway, b) operates within a well-established statistical framework amenable to direct control of false positive or false discovery rates, c) exhibits improved power over widely used methods under normal location-based alternative hypotheses, and d) handles complex experimental designs for which permutation resampling is difficult. We compare the properties of this mixed models approach with nonparametric method GSEA and parametric method PAGE using a simulation study, and illustrate its application with a diabetes data set and a dose-response data set.