Automated de novo Transcriptome Analysis of Eukaryotic Organisms

Advancements in next generation sequencing have enabled researchers to broaden the scope of their research from single genes to whole genomes. The sequencing of the transcriptome, or genome-wide expressed genes, is a powerful, cost-effective method for analyzing unique biological characteristics at the molecular level in non-model organisms. De novo transcriptome assembly programs piece together millions of short RNA or DNA sequences into contigs that represent the genes present in the organism of interest. The purpose of this study was to automate a de novo transcriptomics pipeline for eukaryotic organisms. This pipeline was developed using the DoIt python package on a Redhat Linux server at the University of California-San Diego. Following quality trimming and removing subsumed reads, the RNA-Seq reads were assembled with Trinity, a de novo transcriptome assembly program that is able to resolve genes to the isoform level. Each task in the assembly pipeline was iteratively tested, and the resulting assemblies were evaluated using Trinity statistical metrics and CEGMA. Extensive testing with varying polymorphic eukaryotic organisms, including several marine species, helped to optimize this automated pipeline for the pre-processing, assembly, annotation, and validation of transcriptomes. With this automation tool, biologists lacking bioinformatics training will be able to easily assemble a de novo transcriptome for any eukaryotic organism.