The statistical inference of function through evolutionary relationships (SIFTER) tool bases its analysis of protein function on protein family evolution. Developed by the 'Functional annotation of microbial communities in human health and in the environment' (SIFTERMETAGENOMICS) project, the model bases predictions of protein function on different features of the protein evolutionary family tree. Factors taken into account include potential evolution to another function in the family, estimated mutability of the pair of molecular functions, tree distance and the event behind function — speciation or gene duplication. The Critical Assessment of Function protein Annotation (CAFA) results show that SIFTER is in the top 10 performing algorithms, as published in "Nature Methods" in 2013. Not only that, but the tool showed its potential in a study on Crohn's disease. It is now widely accepted that Crohn's disease is associated with an imbalance in the gut microbial ecosystem. The project researchers performed 16S ribosomal profiling of Crohn's disease patients and a control population of patients (both groups were undergoing surgery or colonoscopy). The profiles of Crohn's disease patients showed that the microbiota of those in remission were more similar to the control group than patients who suffered a recurrence of the disease. Furthermore, patients who remained in remission showed greater stability of gut microbial proteins. SIFTERMETAGENOMICS protein analyses demonstrated that profiling of gut microbiota could be useful in predicting treatment efficacy in Crohn's disease patients undergoing surgery. Applications for the SIFTER tool are in line with initiatives such as the 'Critical assessment of genome interpretation' (CAGI). A community experiment, CAGI is used to assess computational methods for predicting the phenotypic impacts of genomic variation. The CAGI experiment is ideally suited to the analysis of Crohn's disease data. For the latest conference there were 56 submissions for this condition. In all, results from CAGI and projects like SIFTERMETAGENOMICS will help clinicians and counsellors understand the reliability of variant prediction methods and their appropriate application.
Functional annotation of microbial communities in human health and in the environment
Discover other articles in the same domain of application
29 March 2018