Transcriptogramer: pacote em R para análise transcricional

Abstract

The transcriptogram, a method used on transcriptomes analysis, uses protein-protein interaction data to build an ordered gene list. On this list, genes are placed such that the probability of interaction between its products exponentially decreases with the increase of the distance between its positions. The ordered gene list is then used to calculate the average expression value of functionally associated genes in a window with settable radius, allowing the differential expression of non-predefined gene sets in case-control studies. This study aims to implement an R package that uses transcriptograms and integrates features from packages known by the scientific community, able to perform: differential expression, functional enrichment, and network visualization. The transcriptogramer package was implemented and is available at Bioconductor, a repository for open source softwares developed in the R language for use in bioinformatics. In a comparison between the transcriptogramer and a pipeline combining features from limma and topGO packages, was noticed that the transcriptogramer identified nearly 10 times more Gene Ontology terms significantly enriched, among which most of the terms identified by the conventional pipeline were found.