Overview of metatranscriptome simulation pipeline based on FluxSimulator. For each species considered, the genome sequence and ORF annotation file is used to create a list of predicted transcripts for each species. FluxSimulator then assigns each gene a random expression value based on Zipf’s law to create a library of the mRNA molecules that are present in the sample. Given a list of experimental parameter input (sequence errors, sample bias, and relative species abundance), a set of simulated metatranscriptomic reads are generated based on the set of transcripts provided. A gold standard assembly is then generated by aligning reads to the original transcripts and obtaining consensus sequences from the resulting alignments.