You are here:
Gene prediction in the fathead minnow [Pimephales promelas] genome-presentation
Citation:
Saari, T., Dan Villeneuve, G. Ankley, F. Burns, A. Cogburn, S. Deschamps, R. Jackson, R. Hoke, AND A. Schroeder. Gene prediction in the fathead minnow [Pimephales promelas] genome-presentation. SETAC, North America, Salt Lake City, UT, November 01 - 05, 2015.
Impact/Purpose:
not applicable
Description:
The fathead minnow is a well-established ecotoxicological model organism, having been widely used for regulatory ecotoxicity testing and research for over a half century. While a large amount of molecular information has been gathered on the organism over the years, to date, the fathead minnow genome has been largely uncharacterized. To address this limitation, high-throughput Illumina technology was recently employed to sequence the fathead minnow genome. Approximately 100X coverage was achieved by sequencing several libraries of paired-end reads with differing genome insert sizes. Two draft assemblies were generated using the SOAPdenovo and String Graph Assembler (SGA) methods. Mapping a set of highly conserved eukaryotic genes to each assembly revealed the SOAPdenovo assembly as adequate for initiating the gene prediction and annotation process, with 91% of genes mapping completely, and 74% entirely on one scaffold. The gene prediction process was initiated using the ab initio capabilities of the program AUGUSTUS. Different training sets of gene structures were tested to optimize ab initio gene prediction accuracy. Of the training sets tested, a recently-released species model for zebrafish performed the best, delivering an exon-level sensitivity and specificity of 65.6% and 72.7%, respectively. Using this model, 44,401 potential coding sequences were found within the fathead minnow genome. A second iteration of gene prediction will make use of extrinsic gene information, such as fathead minnow expressed sequence tags (EST) and protein alignments, to improve accuracy. Over 20,000 EST clusters and have been aligned to the fathead minnow genome using exonerate. The ongoing process of annotation will provide further molecular characterization useful for future design of ecotoxicology studies using the fathead minnow.