Maximum Entropy Conditional Probability Modeling
This new paper by Miller et al. looks interesting.
Miller DJ, Zhang Y, Yu G, Liu Y, Chen L, Langefeld CD, Herrington D, Wang Y. An Algorithm for Learning Maximum Entropy Probability Models of Disease Risk That Efficiently Searches and Sparingly Encodes Multilocus Genomic Interactions. Bioinformatics, in press, 2009. [PubMed]
MOTIVATION: In both genome-wide association studies (GWAS) and pathway analysis, the modest sample size relative to the number of genetic markers presents formidable computational, statistical, and methodological challenges for accurately identifying markers/interactions and for building phenotype-predictive models. RESULTS: We address these objectives via maximum entropy conditional probability modeling (MECPM), coupled with a novel model structure search. Unlike neural networks and support vector machines (SVMs), MECPM makes explicit and is determined by the interactions that confer phenotype-predictive power. Our method identifies both a marker subset and the multiple k-way interactions between these markers. Additional key aspects are: i) evaluation of a select subset of up to 5-way interactions while retaining relatively low complexity; ii) flexible SNP coding (dominant, recessive) within each interaction; iii) no mathematical interaction form assumed; iv) model structure and order selection based on the Bayesian Information Criterion, which fairly compares interactions at different orders and automatically sets the experiment-wide significance level; v) MECPM directly yields a phenotype-predictive model. MECPM was compared to a panel of methods on data sets with up to 1000 SNPs and up to 8 embedded penetrance function (i.e., ground-truth) interactions, including a 5- way, involving less than 20 SNPs. MECPM achieved improved sensitivity and specificity for detecting both ground-truth markers and interactions, compared with previous methods. AVAILABILITY: http://www.cbil.ece.vt.edu/ResearchOngoingSNP.htm CONTACT: firstname.lastname@example.org.