Abstract 3273: rkmh: A MinHash toolbox for analyzing HPV coinfections

Abstract

Abstract Human papillomaviruses are common infections transmitted by direct physical contact. Two sexually transmitted types, HPV16 and HPV18, are responsible for almost 75% of cervical cancer cases worldwide; another 10 types of >200 known types are considered carcinogenic. Among the four HPV16 lineages (A, B, C, and D), 100-fold differences in odds ratios for adenocarcinoma are observed. It is therefore informative to distinguish between infection with HPV16 lineages, even though they differ by as little as 2% of their genome. We present a computational toolkit, rkmh, for characterizing viral coinfections. rkmh uses kmer matching strategies to rapidly determine the most similar type or lineage reference genome for a given read. The proportion of reads matching each reference genome can then be calculated and the ratios of the infecting viruses can be estimated. To assess the performance of rkmh we first simulated 100bp paired end Illumina read sets from the PAVE database of HPV reference genomes, then also evaluated performance on a real HPV16 sample sequenced on the Ion Torrent Proton platform (typical read length 250bp), and a set of 3,660 Oxford Nanopore minION reads generated from two HPV16 reference strains (typical read length over 6500bp). We demonstrate that rkmh can adequately classify HPV infections at the type and lineage level. We discuss further applications of the tool in metagenomics. rkmh is freely available at https://github.com/edawson/rkmh. Citation Format: Eric T. Dawson, Sarah Wagner, David Roberson, Meredith Yeager, Joseph Boland, Erik Garrison, Mark Schiffman, Tina Raine-Bennet, Thomas Lorey, Phillip Castle, Stephen Chanock, Lisa Mirabello, Richard Durbin. rkmh: A MinHash toolbox for analyzing HPV coinfections [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3273.