In this study, researchers used a single-cell strategy to analyze and interpret deoxyribonucleic acid from multiple donors.
This study introduces a novel single-cell strategy to analyze and interpret deoxyribonucleic acid from multiple donors, demonstrating the feasibility and statistical robustness of such a strategy. It provides an integrated approach for the probabilistic interpretation of forensic DNA data from multiple donors, improving both the efficiency and reliability of forensic DNA analysis. Researchers used metrics for electropherogram (EPG) signal quality and statistical tests to optimize the process from extraction to interpretation. The method used combines multiple approaches of direct-to-PCR extraction, microfluidic DEPArray™ technology, and Model-Based Clustering (MBC) algorithms to improve the reliability and interpretability of forensic DNA analysis. The researchers evaluated four direct-to-PCR extraction treatments across 102 single buccal cells, measuring allele detection rates, peak heights, peak height ratios, and peak height balance for short tandem repeat (STR) markers. Statistical tests identified significant EPG metric variations among treatments. The Arcturus® PicoPure™ extraction method showed the most efficiency, exhibiting the lowest median allele drop-out rate and the highest median average peak height, among others. The researchers adjusted reagent concentrations, specifically phosphate buffer saline (PBS) and proteinase K. Results indicated that decreased PBS concentrations from 1X to 0.5X or 0.25X significantly improved EPG quality metrics, while variation of proteinase K concentrations yielded negligible impact. Leveraging DEPArray™ technology, the research further incorporated a semi-automated single-cell strategy optimized for the forensic pipeline. An innovative framework was developed to streamline the interpretation of single-cell electropherograms (scEPGs) using MBC and a novel algorithm, EESCIt (Evidentiary Evaluation of Single Cells). This framework clustered scEPGs by genetic origin calculates likelihood ratios (LR) for each cluster and averaged these across all clusters to provide a comprehensive weight of evidence summary. Analysis revealed that 99% of comparisons provided log LR values greater than 0 for true contributors, regardless of mixture complexity. It significantly expanded the applicability of DNA forensics.