PerSVade: personalized structural variant detection in any species of interest
Visualitza/Obre
Cita com:
hdl:2117/372015
Tipus de documentArticle
Data publicació2022
EditorBioMed Central (BMC)
Condicions d'accésAccés obert
Llevat que s'hi indiqui el contrari, els
continguts d'aquesta obra estan subjectes a la llicència de Creative Commons
:
Reconeixement 4.0 Internacional
ProjecteEVOLUCION GENOMICA Y FENOTIPICA EN PATOGENOS DEL GENERO CANDIDA (AEI-PGC2018-099921-B-I00)
RETVOLUTION - Reticulate evolution: patterns and impacts of non-vertical inheritance in eukaryotic genomes. (EC-H2020-724173)
RETVOLUTION - Reticulate evolution: patterns and impacts of non-vertical inheritance in eukaryotic genomes. (EC-H2020-724173)
Abstract
Structural variants (SVs) underlie genomic variation but are often overlooked due to difficult detection from short reads. Most algorithms have been tested on humans, and it remains unclear how applicable they are in other organisms. To solve this, we develop perSVade (personalized structural variation detection), a sample-tailored pipeline that provides optimally called SVs and their inferred accuracy, as well as small and copy number variants. PerSVade increases SV calling accuracy on a benchmark of six eukaryotes. We find no universal set of optimal parameters, underscoring the need for sample-specific parameter optimization. PerSVade will facilitate SV detection and study across diverse organisms.
DatasetPerSVade is available at https://github.com/Gabaldonlab/perSVade [39] and can be installed using either conda environments or through a docker image containing the pipeline, available at https://hub.docker.com/r/mikischikora/persvade. The github repository is released under an open source GNU General Public License (GPL). In addition, the code can be accessed in Zenodo through the DOI 10.5281/zenodo.6866529 [86]. The github repository contains detailed examples on how to install and run perSVade using conda, docker, or singularity. We have tested perSVade on several Linux and Mac architectures, and the docker image may be run in any machine in a reproducible way. All the results shown in this paper were generated using the script https://github.com/Gabaldonlab/perSVade/blob/master/scripts/perSVade.py from version 1.0, which is a wrapper to execute several modules with a single command. Since perSVade is an actively used (and maintained) pipeline, we have created a few new releases since version 1.0, which include an improved documentation, more unit tests, and the implementation of an efficient debugging of inputs. Note that these changes do not affect the functionality of the modules as implemented in version 1.0. Hence, we recommend the usage of the latest version (version 1.02.7 at the time of publication), which is the one with the best documentation and usability. In addition, note that this one-liner wrapper is not recommended for broad usage. All the data used for testing perSVade was obtained from the SRA database or public ftp servers, and is listed in Additional file 1: Table S1 and “Methods.” All the code necessary to reproduce the results and plots shown in this paper is in https://github.com/Gabaldonlab/perSVade/tree/master/testing.
CitacióSchikora Tamarit, M.À.; Gabaldón, T. PerSVade: personalized structural variant detection in any species of interest. "Genome Biology", 2022, vol. 23, 175.
ISSN1474-760X
1465-6906
1465-6906
Col·leccions
Fitxers | Descripció | Mida | Format | Visualitza |
---|---|---|---|---|
s13059-022-02737-4.pdf | 4,730Mb | Visualitza/Obre |