Assessment of High Throughput Sequencing Quality of Host <scp>DNA</scp> Enriched From Faeces: A Case From Captive Tiger

Faeces serve as a widely used non-invasive material in wildlife studies, from which host DNA can be enriched for high-throughput sequencing. However, the characteristics of faecal DNA sequencing, particularly the relationship between sequencing volume and genotyping accuracy or genomic coverage, have remained unclear. This knowledge gap makes it challenging to pre-estimate the required sequencing depth to ensure data quality while controlling costs. In this study, we used host-enriched faecal DNA obtained via peri-extraction enrichment by SDS (PEERS) to perform comparative analyses of whole-genome sequencing data from paired faecal and blood samples of 11 captive Amur tigers. We systematically evaluated how data volume affects genomic coverage and genotyping accuracy. Our results revealed that faecal sequencing reads cover the genome unevenly, which may compromise the reliability of underrepresented allele identification and subsequent genomic analyses. Specifically, genomic coverage and genotyping accuracy, including precision, error rate, recall and F1-score, improved with increasing sequencing depth in both blood and faecal DNA. For reliable genotyping (values > 0.95), a minimum average depth of 20× was required for blood-derived DNA, whereas faecal DNA needed up to 50× depth when the nuclear DNA enrichment efficiency (EE_N, defined as the ratio of nuclear to bacterial DNA copy number) reached 5E-5. At this sequencing volume, population-level indices, such as nucleotide diversity (π), polymorphism information content (PIC), the distribution of π in 500-kb sliding windows and SNP density, as well as individual-level indices, including heterozygosity, kinship coefficients, inbreeding coefficient (F_ROH) and genetic load, were highly consistent between faecal and blood samples. Only observed heterozygosity (Ho) and kernel density distributions of minor allele frequency (MAF) showed slight discrepancies. These findings underscore that effective host DNA enrichment using PEERS combined with deep sequencing is essential to achieve adequate genomic coverage and accurate genotyping when utilising faecal samples. Monitoring enrichment efficiency can help determine the necessary sequencing depth, facilitating cost-effective experimental design. This study offers practical guidance for the application of faecal DNA in genomic research and supports the broader use of this non-invasive material in future wildlife studies.