RT Journal Article
SR Electronic
T1 TreeToReads - a pipeline for simulating raw reads from phylogenies
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 037655
DO 10.1101/037655
A1 Emily Jane McTavish
A1 James Pettengill
A1 Steve Davis
A1 Hugh Rand
A1 Errol Strain
A1 Marc Allard
A1 Ruth E. Timme
YR 2016
UL http://biorxiv.org/content/early/2016/01/22/037655.abstract
AB Using genome-wide SNP-based methods for tracking pathogens has become standard practice in academia and public health agencies. There are multiple computational approaches available that perform a similar task: call variants by mapping short read data against a reference genome, quality filter these variants, then concatenate the variants into a sequence matrix for downstream phylogenetic analysis. However, there are no existing methods to validate the accuracy of these approaches despite the fact that we know there are parameters that can affect whether a SNP is called, or the correct tree is recovered. We present a simulation approach (TreeToReads) to generate raw read data from mutated genomes simulated under a known phylogeny. The user can vary parameters of interest at each step in the simulation (e.g., topology, model of sequence evolution, and read coverage) to assess the robustness of a given result, which is critical within both research and applied settings. Source code, examples, and a tutorial are available at https://github.com/snacktavish/TreeToReads.