TY - JOUR T1 - High-throughput pipeline for the <em>de novo</em> viral genome assembly and identification of minority variants from Next-Generation Sequencing of residual diagnostic samples JF - bioRxiv DO - 10.1101/035154 SP - 035154 AU - T Gallo Cassarino AU - D Frampton AU - R Sugar AU - E Charles AU - Z Kozlakidis AU - P Kellam Y1 - 2016/01/01 UR - http://biorxiv.org/content/early/2016/05/17/035154.abstract N2 - Motivation The underlying genomic variation of a large number of pathogenic viruses can give rise to drug resistant mutations resulting in treatment failure. Next generation sequencing (NGS) enables the identification of viral quasi-species and the quantification of minority variants in clinical samples; therefore, it can be of direct benefit by detecting drug resistant mutations and devising optimal treatment strategies for individual patients.Results The ICONIC (InfeCtion respONse through vIrus genomiCs) project has developed an automated, portable and customisable high-throughput computational pipeline to assemble de novo whole viral genomes, either segmented or non-segmented, and quantify minority variants using residual diagnostic samples. The pipeline has been benchmarked on a dedicated High-Performance Computing cluster using paired-end reads from RSV, HIV and Influenza clinical samples. The median length of generated genomes was 82% for the HIV dataset and 100% for each Influenza segment. The analysis of each set lasted less than 12 hours; each sample took around 3 hours and required a maximum memory of 10 GB. The pipeline can be easily ported to a dedicated server or cluster through either an installation script or a docker image. As it enables the subtyping of viral samples and the detection of relevant drug resistance mutations within three days of sample collection, our pipeline could operate within existing clinical reporting time frames and potentially be used as a decision support tool towards more effective personalised patient treatments.Availability The software and its documentation are available from https://github.com/ICONIC-UCL/pipelineContact t.cassarino{at}ucl.ac.ukSupplementary information Supplementary data are available at Briefings in Bioinformatics online.Biographical note Tiziano Gallo Cassarino, PhD, is a Research Associate at University College London; Dan Frampton, PhD, is a Research Fellow at University College London; Robert Sugar, PhD, is a software architect at Intel Health and Life Sciences; Elijah Charles is an architect at Intel Health and Life Sciences; Zisis Kozlakidis, PhD, is the ICONIC project manager and the Head of the Centre of Excellence for Infectious Diseases - BBMRI.uk; Paul Kellam, PhD, is the ICONIC principal investigator and professor of Virus Genomics at Imperial College London. ER -