Hair is a common form of biological evidence recovered from crime scenes; however, its forensic utility has historically been limited. Although hair contains both mitochondrial and nuclear DNA, mitochondrial DNA lacks sufficient discriminatory power for individualization, while nuclear DNA in shed (rootless) hairs is typically present in low quantities and is highly degraded. Consequently, traditional STR-based profiling, which requires DNA fragments larger than those typically found in hair shaft, often fails when applied to hair shafts. Genotyping approaches based on single nucleotide polymorphisms (SNPs) offer a promising alternative for analyzing degraded samples. However, their routine application has been limited by challenges in recovering and enriching nuclear DNA from hair shafts.
This study aimed to develop and optimize a robust massively parallel sequencing (MPS) workflow incorporating hybridization-based target enrichment to enable nuclear SNP genotyping from rootless hair shafts. Multiple DNA extraction strategies were evaluated, and a modified BioChain® cfDNA-based method was selected for its efficiency and compatibility with low-input samples. Library preparation was performed using the Takara ThruPLEX® DNA Seq kit. A custom hybridization probe panel targeting 818 SNPs was developed using Daicel Arbor Biosciences technology. Libraries underwent two rounds of hybridization-based target capture prior to sequencing on an Illumina NextSeq platform.
Nuclear DNA was extracted from 90 donor hair samples, with approximately 37% of 5 cm hair segments yielding at least 50 pg of DNA. Twenty-one samples were processed through the complete workflow. Sequencing data were analyzed using CLC Genomics Workbench (Qiagen) to assess fragment size distribution, alignment quality, and locus-level coverage. SNP recovery ranged from 7% to 98% of targeted autosomal loci at a minimum coverage threshold of 10X. Input DNA quantity was correlated with locus recovery with r2=0.78. Fragment lengths (~30–400 bp) confirmed severe degradation, and a ~10.4 bp periodicity consistent with nucleosome-associated degradation was observed.
Genotypes obtained from hair shafts were compared to reference buccal samples using FamLink2 to calculate likelihood ratios (LRs) for identity. In all cases, including low-yield samples, LRs supported inclusion of the true donor and exclusion of non-donors, with values comparable to or exceeding those obtained using conventional STR analysis.
These findings demonstrate that nuclear SNP genotyping from rootless hair shafts is feasible using optimized extraction and target enrichment, enhancing the evidentiary value of hair and supporting broader adoption of MPS-based approaches for challenging forensic samples.