Sexing carnivores from faecal metagenomes: calibrating Rx for stone marten and red fox
- 1Centre for Bioinformatics, University of Veterinary Medicine, Budapest, 1078, Hungary
Abstract
Knowing the sex of the animal that deposited a scat adds value to non-invasive monitoring, but read-ratio sexing methods were developed for human ancient DNA. We examined their use in shotgun metagenomes of seven stone marten (Martes foina) and red fox (Vulpes vulpes) scats. Human DNA, present in every library, was removed by competitive mapping against the host assembly and the human reference. Applying the Skoglund Ry metric to the marten samples yielded 0 to 25 Y-mapping reads per sample and failed to identify either of the two animals that Rx places firmly in the male range, so we adopted the X-to-autosome ratio Rx of Mittnik et al. instead, reimplemented as an open-source, species-agnostic Python tool. To test whether its human-derived thresholds transfer, we simulated female and male genomes of both species and drew 200 replicate subsamples at each of fourteen depths from 1,000 to 100,000 mapped reads. Nine of 11,200 replicates were assigned to the wrong sex, all of them below 2,000 mapped reads; above that depth the only failure was non-assignment, in under 5% of replicates above 7,500 reads and under 1% above 15,000. Applied to the scats, Rx identified four females and two males; one sample with 596 host-assigned reads was left unassigned. With competitive removal of human reads, Rx and its original thresholds are usable in these carnivores.
Related
- Spatiotemporal Overlap and Scent-Marking Behaviour of the European Badger and the Stone Marten in a Mediterranean Landscape.
- Beyond Terrestrial Carnivores: Olfactory Bulb Anatomy in a Semi-Aquatic Mammal
- Bill colour outperforms iris colour for sexing in Starlings Sturnus vulgaris in breeding condition
- Copy Number Variation‐Based Molecular Sexing of Ixodes scapularis and Rhipicephalus microplus Immature Stages Using qPCR and ddPCR Approaches
- Sexing Viking Age horses from burial and non-burial sites in Iceland using ancient DNA
BibTeX
@misc{vujic2026sexing,
title = {{Sexing carnivores from faecal metagenomes: calibrating Rx for stone marten and red fox}},
author = {V. Vujic and C. Mayer and G. Helyes and A. Maroti-Agots and G. Maroti and I. Csabai and N. Solymosi},
howpublished = {bioRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.04.764345},
url = {https://www.biorxiv.org/content/10.64898/2026.11.04.764345v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Sexing carnivores from faecal metagenomes: calibrating Rx for stone marten and red fox
AU - V. Vujic
AU - C. Mayer
AU - G. Helyes
AU - A. Maroti-Agots
AU - G. Maroti
AU - I. Csabai
AU - N. Solymosi
PY - 2026
DA - 2026/11/04
PB - bioRxiv
DO - 10.64898/2026.11.04.764345
UR - https://www.biorxiv.org/content/10.64898/2026.11.04.764345v1
AB - Knowing the sex of the animal that deposited a scat adds value to non-invasive monitoring, but read-ratio sexing methods were developed for human ancient DNA. We examined their use in shotgun metagenomes of seven stone marten (Martes foina) and red fox (Vulpes vulpes) scats. Human DNA, present in every library, was removed by competitive mapping against the host assembly and the human reference. Applying the Skoglund Ry metric to the marten samples yielded 0 to 25 Y-mapping reads per sample and failed to identify either of the two animals that Rx places firmly in the male range, so we adopted the X-to-autosome ratio Rx of Mittnik et al. instead, reimplemented as an open-source, species-agnostic Python tool. To test whether its human-derived thresholds transfer, we simulated female and male genomes of both species and drew 200 replicate subsamples at each of fourteen depths from 1,000 to 100,000 mapped reads. Nine of 11,200 replicates were assigned to the wrong sex, all of them below 2,000 mapped reads; above that depth the only failure was non-assignment, in under 5% of replicates above 7,500 reads and under 1% above 15,000. Applied to the scats, Rx identified four females and two males; one sample with 596 host-assigned reads was left unassigned. With competitive removal of human reads, Rx and its original thresholds are usable in these carnivores.
ER -