Introducing the hermes package for analyzing and reporting RNA-seq data
Daniel Sabanes Bove
The hermes package published on Bioconductor provides classes and functions for quality control, filtering, normalization and differential expression analysis of pre-processed RNA-seq data. Data can be imported from
SummarizedExperiment as well as
matrix objects and can be annotated from BioMart. Filtering for genes without too low expression or containing required annotations, as well as filtering for samples with sufficient correlation to other samples or total number of reads is supported. The standard normalization methods including
tpm can be used, and
DESeq2 as well as
voom differential expression analyses are available. We present the package, as well as a (still closed but soon open source) extension including Shiny modules based on the “teal” framework.