Performance of gene expression analyses usingde novoassembled transcripts in polyploid species

Motivation Quality of gene expression analyses using de novo assembled transcripts in species experienced recent polyploidization is yet unexplored. Results Five plant species with various polyploidy history were used for differential gene expression (DGE) analyses. DGE analyses using putative genes inferred by Trinity performed similar to or better than Corset and Grouper in precision, but lower in sensitivity. In species that lack polyploidy event in the past few million years, DGE analyses using de novo assembled transcriptome identified 50–76% of the differentially expressed genes recovered by mapping reads to the reference genes. However, in species with more recent polyploidy event, the percentage decreased to 7–30%. In addition, 7–89% of differentially expressed genes from de novo assembly are contaminations. Gene co-expression network analyses using de novo assemblies vs. mapping to the reference genes recovered the same module that significantly correlated with treatment in one of the five species tested. Availability and Implementation Commands and scripts used in this study are available at https://bitbucket.org/lychen83/chen_et_al_2018_benchmark_dge/ ; Analysis files are available at Dryad doi: XXXXXX. Contact [email protected] Supplementary information Supplementary data are available at Bioinformatics online