Optimization of Genotype by Sequencing data for phylogenetic purposes

L.O Loureiro,M.D Engstrom,B.K Lim

doi:10.1016/j.mex.2020.100892

Abstract

• Herein we propose a framework for assembling and analyzing Genotype by Sequencing (GBS) data to better understand evolutionary relationships within a group of closely related species using the mastiff bats (Molossus) as our model system. Many species within this genus have low-levels of genetic variation within and between morphologically distinct species, and the relationships among them remain unresolved using traditional Sanger sequencing methods. Given that both de novo and reference genome pipelines can be used to assemble next generation sequences, and that several tree inference methodologies have been proposed for single nucleotide polymorphism (SNP) data, we test whether different alignments and phylogenetic approaches produce similar results. We also examined how the process of SNP identification and mapping can affect the consistency of the analyses. Different alignments and phylogenetic inferences produced consistent results, supporting the GBS approach for answering evolutionary questions on a macroevolutionary scale when the genetic distance among phenotypically identifiable clades is low. We highlight the importance of exploring the relationships among groups using different assembly assumptions and also distinct phylogenetic inference methods, particularly when addressing phylogenetic questions in genetic and morphologically conservative taxa.• The method uses the comparison of several filter settings, alignments, and tree inference approaches on Genotype by Sequencing data.• Consistent results were found among several approaches.• The methodology successfully recovered well supported species boundaries and phylogenetic relationships among species of mastiff bats not hypothesized by previous methods.

Highlights

This technique provides sequence data for thousands of single nucleotide polymorphisms (SNPs), allowing the detection of small, but consistent genetic variation among genetically similar groups not revealed by standard gene sequencing approaches
We propose a framework for assembling and analyzing Genotype by Sequencing (GBS) data to better understand evolutionary relationships among species of mastiff bats (Molossus), a genus with a complex taxonomic history and low levels of genetic variation [7]
We test how four different filtering settings affect the accuracy and consistency of our data. Given that both de novo and reference genome pipelines are often used to assemble generation sequencing (NGS) data, and that several tree inference methodologies have been proposed for SNP data, we test if different alignments and phylogenetic approaches produce similar results

Summary

Background

Advances in genomics technology have allowed the generation of large numbers of molecular markers across the genome, which increases sample sizes and provides additional data to help resolve interpretation of the ecology and evolution of traditionally poorly understood species groups [1]. We test how four different filtering settings affect the accuracy and consistency of our data Given that both de novo and reference genome pipelines are often used to assemble generation sequencing (NGS) data, and that several tree inference methodologies have been proposed for SNP data, we test if different alignments and phylogenetic approaches produce similar results. These data offer a useful framework for other comparative studies of ecology and evolution using the GBS approach

Methodology

Findings

Method validation