Fig. 6

Developed scripts and published programs (i.e. black letters) with their corresponding effects (i.e. grew letters) implemented in the driving script ‘fastGOEA’ performing GOEA based on the parent-child approach integrating hypergeometric tests and Bonferroni corrections. This workflow examines the Uniprot API (i.e. slow mode) or by default a local dataset of gene identifiers (fast mode) from Uniprot (i.e. idmapping.selected.table.gz) in order to associate GO-terms to gene identifiers provided by the user as two lists referring to the sample and universe of hypergeometric tests. With regard to a directed acyclic graph (DAG) of GO-terms (i.e. go-basic.obo including eukaryotic and prokaryotic GO-terms or gosubset_prok.obo including prokaryotic GO-terms), this workflow retains prokaryotic GO-terms and avoids obsolete GO-terms before to perform GOEA. These scripts were written with Python 2.7 and implement R libraries ‘p.ajust’, ‘phyper’ and ‘ggplot2’. The whole workflow is automated and the scripts ‘trim’ and ‘enrich’ has to be performed for each GO-terms during trimming and enrichment steps, respectively