CSAX: Characterizing Systematic Anomalies in eXpression Data
- 1 May 2015
- journal article
- research article
- Published by Mary Ann Liebert Inc in Journal of Computational Biology
- Vol. 22 (5), 402-413
- https://doi.org/10.1089/cmb.2014.0155
Abstract
Methods for translating gene expression signatures into clinically relevant information have typically relied upon having many samples from patients with similar molecular phenotypes. Here, we address the question of what can be done when it is relatively easy to obtain healthy patient samples, but when abnormalities corresponding to disease states may be rare and one-of-a-kind. The associated computational challenge, anomaly detection, is a well-studied machine-learning problem. However, due to the dimensionality and variability of expression data, existing methods based on feature space analysis or individual anomalously expressed genes are insufficient. We present a novel approach, CSAX, that identifies pathways in an individual sample in which the normal expression relationships are disrupted. To evaluate our approach, we have compiled and released a compendium of public expression data sets, reformulated to create a test bed for anomaly detection. We demonstrate the accuracy of CSAX on the data sets in our compendium, compare it to other leading methods, and show that CSAX aids in both identifying anomalies and explaining their underlying biology. We describe an approach to characterizing the difficulty of specific expression anomaly detection tasks. We then illustrate CSAX's value in two developmental case studies. Confirming prior hypotheses, CSAX highlights disruption of platelet activation pathways in a neonate with retinopathy of prematurity and identifies, for the first time, dysregulated oxidative stress response in second trimester amniotic fluid of fetuses with obese mothers. Our approach provides an important step toward identification of individual disease patterns in the era of precision medicine.Keywords
This publication has 56 references indexed in Scilit:
- Thrombocytopenia and retinopathy of prematurityJournal of American Association for Pediatric Ophthalmology and Strabismus, 2011
- FRaC: a feature-modeling approach for semi-supervised and unsupervised anomaly detectionData Mining and Knowledge Discovery, 2011
- Transcriptomic analysis of cell-free fetal RNA suggests a specific molecular phenotype in trisomy 18Human Genetics, 2010
- High fat diet-induced maternal obesity alters fetal hippocampal developmentInternational Journal of Developmental Neuroscience, 2009
- Functional genomic analysis of amniotic fluid cell-free mRNA suggests that oxidative stress is significant in Down syndrome fetusesProceedings of the National Academy of Sciences of the United States of America, 2009
- Src kinase modulates the activation, transport and signalling dynamics of fibroblast growth factor receptorsEMBO Reports, 2007
- Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profilesProceedings of the National Academy of Sciences of the United States of America, 2005
- PGC-1α-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetesNature Genetics, 2003
- Significance analysis of microarrays applied to the ionizing radiation responseProceedings of the National Academy of Sciences of the United States of America, 2001
- Small Sample Issues for Microarray-Based ClassificationComparative and Functional Genomics, 2001