Condel is a method to assess the outcome of nonsynonymous SNVs using a consensus deleteriousness score that combines various tools (e.g. SIFT, Polyphen2, MutationAssessor).

How does it perform?

The combination of methods using Condel outperforms each individual method in the task of classifying mutations as deleterious or neutral.

ROC curve of five individual methods and Condel integrated score

In the figure: pph2, polyphen2; logre, LogR PFam E-value; massess, Mutation Assessor; Condel, Weighed Average Score.


How does it work?

The scores of different methods are weighted using the complementary cumulative distributions produced by the five methods on a dataset of approximately 20000 missense SNPs, both deleterious and neutral. The probability that a predicted deleterious mutation is not a false positive of the method and the probability that a predicted neutral mutation is not a false negative are employed as weights. For details, please read the Condel paper.


How to obtain Condel scores?

Currently, the easiest way to run Condel is using Condel server, which integrates the outputs of SIFT, Polyphen2 and MutationAssessor. Alternatively, it is possible to download a PERL script that computes Condel from five tools. The third option is to obtain a weighted average of the scores of SIFT and Polyphen2 through the Ensembl-variation Variant Effect Predictor, either in its API or webserver form.


Condel-web v. 1.5-SNAPSHOT