DEPArrayTM single-cell technology: A validation study for forensic applications
Janine Schulte,Amke Caliebe,Michael Marciano,Pia Neuschwander,Ilona Seiberle,Eva Scheurer,Iris Schulz
DOI: https://doi.org/10.1016/j.fsigen.2024.103026
IF: 4.453
2024-02-19
Forensic Science International Genetics
Abstract:In forensics investigations, it is common to encounter biological mixtures consisting of homogeneous or heterogeneous components from multiple individuals and with different genetic contributions. One promising mixture deconvolution strategy is the DEPArrayTM technology, which enables the separation of cell populations before genetic analysis. While technological advances are fundamental, their reliable validation is crucial for successful implementation and use for casework. Thus, this study aimed to 1) systematically validate the DEPArrayTM system concerning specificity, sensitivity, repeatability, and contamination occurrences for blood, epithelial, and sperm cells, and 2) evaluate its potential for single-cell analysis in the field of forensic science. Our findings confirmed effective identification of different cell types and the correct assignment of successfully genotyped single cells to their respective donor(s). Using the NGM DetectTM Amplification Kit, the average profile completeness for diploid cells was approximately 80%, with ~ 290 RFUs. In contrast, haploid sperm analysis yielded an average completeness of 51% referring to the haploid reference profile, accompanied by mean peak heights of ~ 176 RFUs. Although certain alleles of heterozygous loci in diploid cells showed strong imbalances, the overall peak balances yielded acceptable values above ≥ 60% with a mean value of 72% ± 0.21, a median of 77%, but with a maximum imbalance of 9% between heterozygous peaks. Locus dropouts were considered stochastic events, exhibiting variations among donors and cell types, with a notable failure incidence observed for TH01. Within the wet-lab experimentation with >500 single cells for the validation, profiling was performed using the consensus approach, where profiles were selected randomly from all data to better mirror real casework results. Nevertheless, complete profiles could be achieved with as few as three diploid cells, while the average success rate increased to 100% when using profiles of 6-10 cells. For sperms, however, a consensus profile with completeness >90% of the autosomal diploid genotype could be attained using ≥15 cells. In addition, the robustness of the consensus approach was evaluated in the absence of the respective reference profile without severe deterioration. Here, increased stutter peaks (≥ 15%) were found as the main artifact in single-cell profiles, while contamination and drop-ins were ascertained as rare events. Lastly, the technique's potential and limitations are discussed, and practical guidance is provided, particularly valuable for cold cases, multiple perpetrator rapes, and analyses of homogeneous mixed evidence.
genetics & heredity,medicine, legal