Metagenomic sequencing of CRISPRs as a new marker to aid in personal identification with low-biomass samples.

Abstract

The high specificity of the human skin microbiome is expected to provide a new marker for personal identification. Metagenomic sequencing of clustered regularly interspaced short palindromic repeats (CRISPRs), which we call metaCRISPR typing, was shown to achieve personal identification accurately. However, the intra-individual variability observed in previous studies, which may be due to poor DNA yields from skin samples, has resulted in non-reproducible results. Furthermore, whether metaCRISPR typing can assist in the forensic human DNA analysis of low-biomass samples, from which the information obtained is insufficient, is unknown. In the present study, we sequenced serially diluted control streptococcal CRISPRs cloned into plasmids to determine the minimum copy number required to obtain reproducible results from metaCRISPR typing. We found that at least 102 copies of CRISPRs are necessary to obtain reproducible results. We then analyzed the skin swab samples using both metaCRISPR typing and human DNA typing. When the DNA extracted from the skin swabs was diluted, no information was obtained from six out of eight samples by human DNA typing. On the other hand, beta diversity indices of spacer sequences compared with reference samples were below 0.8 for three out of six samples, for which no information was obtained from human DNA analysis, indicating that the spacers observed in these samples were similar to those in the references. These results indicate that metaCRISPR typing may contribute to the identification of individuals from whom the samples were obtained, even in cases where human DNA yields are insufficient to perform human DNA analysis.IMPORTANCEPrevious studies have developed new personal identification methods utilizing personal differences in the skin microbiome. However, intra-individual diversity of skin microbiome may preclude the application of microbiome-based personal identification. Moreover, no study has compared microbiome-based personal identification and practical human DNA analysis. Here, we revealed that the results of metaCRISPR typing, a previously developed microbiome-based personal identification method, are stable if the copy number of the marker gene is sufficient. We then analyzed the skin swab samples using both metaCRISPR typing and human DNA analysis. Our results indicate that metaCRISPR typing may provide additional information for personal identification using low-biomass samples that cannot be used for conventional human DNA analysis.