Poster abstracts
Poster number 41 submitted by David Frankhouser
MetaComb: Meta-Transcriptomic Alignment with a Combined Genome
David E Frankhouser (Biomedical Sciences Graduate Program, The Ohio State University), Max Westphal, Alex Pelletier, Angela Urdaneta. Paige Stump, Peter Shields (Comprehensive Cancer Center, The Ohio State University), Peally Yan (Division of Hematology, The Ohio State University), Huiling He , Albert de la Chapelle (Department of Molecular Virology, Immunology & Medical Genetics, The Ohio State University), Carmine Sonzone (Molecular, Cellular, and Developmental Biology, The Ohio State University), Ralf Bundschuh (Department of Physics, The Ohio State University)
Abstract:
The ability to detect bacterial and viral RNA embedded within the host transcriptome is a growing area of research and clinically important to understand their roles in human cancers. Currently, there are a few tools, one being SURPI, that can quantify metatranscriptomic (MT) reads in human RNA sequencing (RNAseq) data. The computational demands, lack of flexibility in study design, and limited accuracy in bacterial detection of these tools limit their applicability, especially for studies that include clinical trials. MetaComb, a computational method for detecting and quantifying MT reads in RNAseq data is both flexible and devoid of these common limitations.
MetaComb aligns reads from an RNAseq experiment to a combined host+microbial database. This allows reads from homologous sequences to be kept when their best alignment is to the microbial sequence. Reads are aligned using the fast, industry standard RNAseq aligner STAR. MetaComb is the only MT tool that leverages paired-end (PE) sequencing technology. We have demonstrated that PE reads increases the unique non-host MT alignment by two-fold. We were able to affirm the sensitivity and specificity of MetaComb-derived MT reads with BLAST. MetaComb cannot only quantify bacterial RNA content at the species level, but also at the genus or higher taxonomic levels. Quantification at higher taxonomic levels increase the number of included reads by reducing instances of multi-mapping that result from sequence homology between related species and strains. As proof-of-principle we validated MetaComb-derived quantification using real-time quantitative PCR. In summary, MetaComb provides: 1) increased accuracy of microbial RNA detection, 2) a flexible workflow well suited to clinical studies, and 3) the ability to quantify at any taxonomic level. MetaComb therefore allows researchers to identify microbial RNA content with minimal computational and manpower investment.
Keywords: metatranscriptomic, RNA-seq, metatranscriptomic-specific pipeline