Functional Toll-Like Receptor (TLR)2 polymorphisms in the susceptibility to inflammatory bowel disease.

BACKGROUND: The recent genome-wide association studies (GWAS) in inflammatory bowel disease (IBD) suggest significant genetic overlap with complex mycobacterial diseases like tuberculosis or leprosy. TLR variants have previously been linked to susceptibility for mycobacterial diseases. Here we investigated the contribution to IBD risk of two TLR2 polymorphisms, the low-prevalence variant Arg753Gln and the GTn microsatellite repeat polymorphism in intron 2. We studied association with disease, possible correlations with phenotype and gene-gene interactions. METHODOLOGY/PRINCIPAL FINDINGS: We conducted a large study in 843 patients with Crohn's disease, 426 patients with ulcerative colitis and 805 healthy, unrelated controls, all of European origin. Overall, the frequency for carriers of shorter GTn repeats in intron 2 of the TLR2 gene, which have previously been associated with low TLR2 expression and high IL-10 production, was slightly elevated in Crohn's disease and ulcerative colitis compared to healthy controls (16.0% resp. 16.7% vs. 12.8%). The highest frequency of short GTn carriers was noted among IBD patients on anti TNF-alpha therapy. However, none of these differences was significant in the multivariate analysis. The Arg753Gln polymorphism showed no association with any clinical subtype of IBD, including extensive colitis, for which such an association was previously described. We found no association with specific phenotypic disease subgroups. Also, epistasis analysis revealed no significant interactions between the two TLR2 variants and confirmed IBD susceptibility genes. CONCLUSIONS: The two functional relevant polymorphisms in TLR2, the GTn microsatellite repeat polymorphism in intron 2 and the Arg753Gln variant do not seem to play a role in the susceptibility to Crohn's disease or ulcerative colitis.

Crowdsourced assessment of common genetic contribution to predicting anti-TNF treatment response in rheumatoid arthritis.

Rheumatoid arthritis (RA) affects millions world-wide. While anti-TNF treatment is widely used to reduce disease progression, treatment fails in ∼one-third of patients. No biomarker currently exists that identifies non-responders before treatment. A rigorous community-based assessment of the utility of SNP data for predicting anti-TNF treatment efficacy in RA patients was performed in the context of a DREAM Challenge (http://www.synapse.org/RA_Challenge). An open challenge framework enabled the comparative evaluation of predictions developed by 73 research groups using the most comprehensive available data and covering a wide range of state-of-the-art modelling methodologies. Despite a significant genetic heritability estimate of treatment non-response trait (h(2)=0.18, P value=0.02), no significant genetic contribution to prediction accuracy is observed. Results formally confirm the expectations of the rheumatology community that SNP information does not significantly improve predictive performance relative to standard clinical traits, thereby justifying a refocusing of future efforts on collection of other data.

MULTITASK FEATURE SELECTION WITH TASK DESCRIPTORS.

Machine learning applications in precision medicine are severely limited by the scarcity of data to learn from. Indeed, training data often contains many more features than samples. To alleviate the resulting statistical issues, the multitask learning framework proposes to learn different but related tasks jointly, rather than independently, by sharing information between these tasks. Within this framework, the joint regularization of model parameters results in models with few non-zero coefficients and that share similar sparsity patterns. We propose a new regularized multitask approach that incorporates task descriptors, hence modulating the amount of information shared between tasks according to their similarity. We show on simulated data that this method outperforms other multitask feature selection approaches, particularly in the case of scarce data. In addition, we demonstrate on peptide MHC-I binding data the ability of the proposed approach to make predictions for new tasks for which no training data is available.