The BitSeq package is targeted for transcript expression analysis and differential expression analysis of RNA-seq data in two stage process. In the first stage it uses Bayesian inference methodology to infer expression of individual transcripts from individual RNA-seq experiments. The second stage of BitSeq embraces the differential expression analysis of transcript expression. Providing expression estimates from replicates of multiple conditions, Log-Normal model of the estimates is used for inferring the condition mean transcript expression and ranking the transcripts based on the likelihood of differential expression.