Entering edit mode
9.2 years ago
Pas
▴
30
Hi all,
I inferred a gene network using ARACNe algorithm from a large set of gene expression data (~500) generated by microarray technology. I would like to assess the performance of the ARACNe algorithm on my data set. To this end I would like to evaluate the positive predictive value (PPV) and other measures ( true positive, false positive etc..) using already established (validated) gene networks from databases like IntAct or KEGG. Do you know if exist a package that allows to perform this kind of analysis?
As far as I know ARACNE infers gene regulatory networks so you should try to compare your results with known regulatory interactions. Use protein-protein interactions or pathway/processes memberships only if you're interested in how your inferred network can predict protein-protein interactions or gene function.
Hi Jean,
many thanks. Yes, I agree with you. I'd like to know if exists a tool ( like an r package) that does this kind of comparison.