PISAD: reference-free intraspecies sample anomalies detection tool based on k-mer counting

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Genomic sequencing involves analyzing diverse data types, risking sample mislabeling.
• Current methods need population references, which are unavailable for many species.
• A robust quality control tool is crucial for expanding genomic research to new species.

Purpose of the Study:

• To develop a reference-free tool for validating sample identities in whole-genome sequencing data.
• To address the limitations of existing methods for nonmodel organisms.
• To improve the accuracy and reliability of genomic data analysis.

Main Methods:

• Developed PISAD (Phased Intraspecies Sample Anomalies Detection), a novel two-stage approach.
• Stage 1: Reference-free single nucleotide polymorphism (SNP) calling to create a variant sketch from low-error data.
• Stage 2: Assessing sample concordance against the variant sketch to verify relationships.

Main Results:

• PISAD effectively validates sample identities across multiple species, including Homo sapiens, Bos taurus, Gallus gallus, Arctia plantaginis, and Pyrus species.
• The tool requires significantly lower data coverage (0.5×) compared to reference-based methods.
• Demonstrated broad applicability to various diploid species.

Conclusions:

• PISAD provides a robust, reference-free solution for sample identity validation in whole-genome sequencing.
• The tool overcomes limitations of existing methods, particularly for nonmodel organisms.
• PISAD enhances the reliability of genomic research by minimizing sample swap errors.