Application of NGS-Generated SNP Data to Complex Crops Studies: The Example of Musa spp.

Over recent years, Next-Generation Sequencing (NGS) has become widely used by scientists to obtain high resolution genomic data from many species. In parallel, different Next-Generation genotyping techniques have been developed. However, determining their respective suitability to different research contexts can be challenging, especially when studying highly heterozygous and/or allopolyploid plants. Genotyping data was generated from diploids and triploids Musa (banana) samples using two restriction-site associated DNA sequencing methods: Genotyping by Sequencing (GBS) and Restriction site Associated DNA markers (RAD). Both methods reduce the complexity of the targeted genomes and allow obtaining millions of markers across many individuals at a reasonable cost. The potential and limitations of both techniques have been highlighted based on different concrete applications such as: i) Filtering of raw data to get highly reliable markers for Genome Wide Association Studies (GWAS), ii) Comparison of the number, depth and distribution of markers obtained from each technique using 12 diploid and 4 triploid accessions, iii) Computation of phylogenetic trees for genetic diversity analyses involving taxonomically distant taxa, and iv) Use of low quality DNA from leaf samples gathered during collecting missions. Even though GBS displays significantly larger read depths, RAD allows a better coverage along the genome. Both techniques have assets and drawbacks and should be chosen according to the application targeted.