Discussion
Our results demonstrate that the FASTQ-formatted data routinely
generated by Diversity Arrays Technology (DArT) as an intermediate step
in their workflow can reliably be used to determine the sex of samples
from non-model organisms, confirming or replacing field-based sex
identification and eliminating the need for additional costly laboratory
sexing analyses. Importantly, a reference genome from the species of
interest does not appear to be necessary, as we obtained robust results
by mapping our data to the reference assembly for the house mouse
(Mus musculus ), which shared a common ancestor withL. conditor 10 million years ago (Steppan & Schenk, 2017). While
the house mouse genome is assembled to the chromosome-level, making
identification of reads mapping to the X chromosome straightforward,
this approach should also work with scaffold-level reference assemblies.
Indeed, Gower et al. (2019) identified X-linked scaffolds in the polar
bear genome (UrsMar1.0) by first mapping all scaffolds against the
chromosome-level dog reference assembly (CanFam3.1), then appliedsexassign to shotgun sequencing data from a third species –
brown bears (Ursus arctos ) – that they mapped to the
putative polar bear X-linked scaffolds. Given that scaffold-level
assemblies are increasingly available for a wide range of taxa, our
results suggest that most DArT end-users working on mammals (or indeed
any diploid organism with a heterogametic sex-determination system)
should be able use their FASTQ data to determine the sex of their
samples.