MOTIVATION: Recently, gene-coexpression relationships have been found to be often conditional and dynamic. Many studies have suggested that single nucleotide polymorphisms (SNPs) have impacts on gene expression variations in human populations. RESULTS: The SNPxGE(2) database contains the computationally predicted human SNP-coexpression associations, i.e. the differential coexpression between two genes is associated with the genotypes of an SNP. These data were generated from a large-scale association study that was based on the HapMap phase I data, which covered 269 individuals from 4 human populations, 556 873 SNPs and 15 000 gene expression profiles. In order to reduce the computational cost, the SNP-coexpression associations were assessed using gap/substitution models, proven to have a comparable power to logistic regression models. The results, at a false discovery rate (FDR) cutoff of 0.1, consisted of 44 769 and 50 792 SNP-coexpression associations based on single and pooled populations, respectively, and can be queried in the SNPxGE(2) database via either gene symbol or reference SNP ID. For each reported association, a detailed information page is provided. AVAILABILITY: http://lambchop.ads.uga.edu/snpxge2/index.php CONTACT: [email protected], [email protected].

MOTIVATION: An accurate diagnostic and prediction will not be achieved unless the disease subtype status for every training sample used in the supervised learning step is accurately known. Such an assumption requires the existence of a perfect tool for disease diagnostic and classification, which is seldom available in the majority of the cases. Thus, the supervised learning step has to be conducted with a statistical model that contemplates and handles potential mislabeling in the input data. RESULTS: A procedure for handling potential mislabeling among training samples in the prediction of disease subtypes using gene expression data was proposed. A real data-based simulation study about the estrogen receptor status (ER+/ER-) of breast cancer patients was conducted. The results demonstrated that when 1-4 training samples (N = 30) were artificially mislabeled, the proposed method was able not only in correcting the ER status of mislabeled training samples but also more importantly in predicting the ER status of validation samples as well as using 'true' training data.