Graham M Hughes1, Emma C Teeling1. 1. School of Biology and Environmental Science, University College Dublin, Dublin 4, Ireland.
Abstract
SUMMARY: A number of limiting factors mean that traditional genome annotation tools either fail or perform sub-optimally when trying to detect coding sequences in poor quality genome assemblies/genome reports. This means that potentially useful data is accessible only to those with specific skills and expertise in assembly and annotation. We present an Assembled-Genome mIning pipeLinE (AGILE) written in Perl that combines bioinformatics tools with a number of steps to overcome the limitations imposed by such assemblies when applied to highly fragmented genomes. Our methodology uses user-specified query genes from a closely related species to mine and annotate coding sequences that would traditionally be missed by standard annotation packages. Despite a focus on mammalian genomes, the generalized implementation means that it may be applied to any genome assembly, providing a means for non-specialists to gather gene sequences for downstream analyses. AVAILABILITY AND IMPLEMENTATION: Source code and associated files are available at: https://github.com/batlabucd/GenomeMining and https://bitbucket.org/BatlabUCD/genomemining/src. Singularity and Virtual Box images available at https://figshare.com/s/a0004bf93dc43484b0c0. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
SUMMARY: A number of limiting factors mean that traditional genome annotation tools either fail or perform sub-optimally when trying to detect coding sequences in poor quality genome assemblies/genome reports. This means that potentially useful data is accessible only to those with specific skills and expertise in assembly and annotation. We present an Assembled-Genome mIning pipeLinE (AGILE) written in Perl that combines bioinformatics tools with a number of steps to overcome the limitations imposed by such assemblies when applied to highly fragmented genomes. Our methodology uses user-specified query genes from a closely related species to mine and annotate coding sequences that would traditionally be missed by standard annotation packages. Despite a focus on mammalian genomes, the generalized implementation means that it may be applied to any genome assembly, providing a means for non-specialists to gather gene sequences for downstream analyses. AVAILABILITY AND IMPLEMENTATION: Source code and associated files are available at: https://github.com/batlabucd/GenomeMining and https://bitbucket.org/BatlabUCD/genomemining/src. Singularity and Virtual Box images available at https://figshare.com/s/a0004bf93dc43484b0c0. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
Authors: Joanna Kacprzyk; Andrea G Locatelli; Graham M Hughes; Zixia Huang; Michael Clarke; Vera Gorbunova; Carlotta Sacchi; Gavin S Stewart; Emma C Teeling Journal: Aging (Albany NY) Date: 2021-03-21 Impact factor: 5.682