CAGI Challenge Gauges Computational Predictions of Phenotype

By Bio-IT World Staff 

September 20, 2011 | A host of deadlines are approaching for the CAGI challenges, starting September 30 and running through the end of the year.  

The Critical Assessment of Genome Interpretation project is a community experiment to objectively assess computational methods for predicting the phenotypic impacts of genomic variation. The experiment is modeled on Critical Assessment of Structure Prediction (CASP); participants will be provided genetic variants and will make predictions of resulting molecular, cellular, or organismal phenotype. These predictions will be evaluated against experimental characterizations. Assessors include Rui Chen, Iddo Friedberg, Gad Getz, Sean Mooney, Pauline Ng, and Sean Tavtigian.

The goal CAGI is to identify bottlenecks in genome interpretation, inform critical areas of future research, and connect researchers from diverse disciplines whose expertise is essential to methods for genome interpretation.  

The datasets used for the challenges include:   

• Disease-associated variants of a human metabolic enzyme. 
• Variants from the resequencing of breast cancer patients and control subjects. 
• Disease-associated variants of a human sodium channel. 
• Genome and RNA-seq data from identical twins with discordant disease. 
• Multiple genomics data for cancer cell lines, with differential response to drugs. 
• Exomes of Crohn’s disease patients and healthy individuals. 
• Double mutants of p53 to identify mutations that restore the activity of inactive p53. 
• Predicting the medical phenotypes of individuals with genome data. 
• Microbial dataset measuring the effect of gene disruptions under stress conditions. 
• The riskSNPs dataset to identify potential causative SNPs from lists of candidates in the disease-associated loci for seven complex trait diseases.