Solution to a case involving the interpretation of trace degraded DNA mixtures
- Open Access
- 07.08.2024
- Case Report
Erschienen in:
Abstract
Introduction
DNA mixture interpretation has long posed significant challenges in forensic genetics. The identification of a suspect’s DNA within specific mixtures can crucially influence the outcome of a case. Trace DNA samples, particularly those from long ago, present increased difficulties due to degradation [1]. Therefore, selecting appropriate genetic markers is essential for the interpretation of degraded trace DNA mixture.
Conventional capillary electrophoresis-based short tandem repeat (CE-STR) is the most widely used method for DNA identification [2]. However, its efficiency in DNA mixture interpretation is limited by factors such as stutter peak generation, allele superposition, allele drop-out from minor contributors, and random amplification effects [3]. These limitations hinder the accurate interpretation of complex DNA mixtures.
Anzeige
In contrast, new genetic markers like microhaplotypes (MHs) have demonstrated significant value in DNA mixture deconvolution [4‐7]. Microhaplotypes, defined as assemblies of single nucleotide polymorphisms (SNPs) less than 300 bp long and detected by next-generation sequencing (NGS) [8, 9], offer advantages over STRs. MHs are free from stutter artifacts and benefit from large multiplex capacities on the NGS platform, making them more effective for DNA mixture interpretation [7]. However, the molecular sizes of reported MHs often exceed 100 bp [5, 10], while the highly-degraded DNA fragments are frequently shorter than 100 bp [11]. Long amplicons may hinder the detection of degraded DNA samples [11]. Although smaller MHs (< 70 bp) have been reported, their number in these studies was too low to provide convincing discrimination [12, 13].
In this case report, we utilized a novel genetic marker akin to MH— multi-SNP (MNP), which combines two or more SNP loci within a 75 bp based on the diversity values. This marker was employed to interpret trace-amount degraded DNA mixtures, contributing to the resolution of a cold case.
Materials and methods
DNA acquisition
The evidence in this case comprised a campstool belonging to the victim, which had been stored in the police office for over a decade. Our objectives were to search for and identify the suspect’s DNA on the campstool. The campstool consisted of eight wooden bars, each of which we divided into three sections — left, middle, and right — for sampling purposes. In total, we obtained sampling swabs from 24 sections of the campstool, as illustrated in Fig. 1. We also collected the suspect’s blood sample and extracted DNA from all the samples using the QIAamp DNA Investigator kit (Qiagen, German). Concurrently, we retrieved the victim’s STR genotypes from a previous testing report.
CE-STR typing
Conventional CE-STR typing was performed to analyze all the DNA samples. STR profiles were determined with the Identifiler Plus PCR amplification kit (Applied Biosystems, USA). Fluorescence multiplex PCR was carried out in accordance with the manufacturer’s instructions, and the amplified products were separated on 3130 XL ABI Prism Genetic Analyzer (Applied Biosystems, USA). Genotyping was then carried out with GeneMapper software (Applied Biosystems, USA). All results were reviewed and verified by two experienced technicians.
Anzeige
Deconvolution of the mixed STR profiles
The GenoProof Mixture 3 (GPM 3) software [14] was applied to deconvolute the mixed STR profiles. The number of contributors was set to two, with the victim’s STR profiles serving as the reference.
MNP sequencing
The ‘FD multi-SNP Mixture Kit’ (Fudan University, China) was used for the multiplex amplification of MNP loci from targeted DNA samples. This kit covers 567 MNPs with high polymorphism and discrimination within the Chinese population, including 48 single SNPs, 157 two-linked SNPs, 246 three-linked SNPs, 91 four-linked SNPs, and 25 five-linked SNPs.
The MGIEasy Library Prep Kit (Beijing Genomics Institute, China) was used to construct a sequencing library and add eight-nucleotide barcode sequences specific to each sample for multiplexing during sequencing. The pooled libraries were sequenced on the Illumina MiSeq platform (Illumina, USA), and the raw data were analyzed using MiSeq Reporter software (Illumina, USA) with default settings. Sequencing reads were then compared to the human reference genome Hg19 using bowtie2 (version 2.1.0) [15] to filter out incomplete reads. For each fully mapped read, the nucleotide sequence spanning a multi-SNP locus was identified as its allele.
Mixed MNP genotypes analysis
To analyze the mixed MNP genotypes, we first determined the number of contributors (\(\:n\)) in the mixed sample. The occurrence of an allele in a mixed sample aligns with its frequency in the population, and the number of occurrences follows a polynomial distribution based on twice the number of contributors in the mixed sample. According to the polynomial distribution, we calculated the probability \(\:P(i{k}_{i};n)\) of observing \(\:k\) allelotypes in the ith marker (\(\:{l}_{i}\)) of the mixed sample consisting of \(\:n\) individuals. Using \(\:P(i{k}_{i};n)\), we derived the likelihood function of the mixed sample in terms of the contributors as \(\:L\left(n|{k}_{1},{k}_{2},{\dots\:,k}_{l}\right)=P\left({k}_{1},{k}_{2},{\dots\:,k}_{l};n\right)=\prod\:_{i=1}^{l}P(i{k}_{i};n).\) The number of contributors in the mixture sample was determined by maximizing the value of this likelihood function.
Subsequently, we evaluated the probability of the suspect’s DNA presence in the mixed sample using the principle of ‘non-splitting’. The calculation method was as follows: upon identifying the suspect’s allele typing in the mixed sample, we formed two hypotheses—one suggesting a random match and the other positing the suspect as a contributor. Utilizing the determined number ‘\(\:n\)’, the probability that the suspect’s allele typing matches the mixed sample DNA is \(\:{P=\left(1-r\right)}^{2n}\), where ‘\(\:r\)’ represents the frequency of that allele typing in the population. Given that ‘\(\:P\)’ adheres to a polynomial distribution, we derived the probabilities of the suspect’s allele types that either match or do not match the DNA of the mixed sample. The resultant probability across all loci was then compared against a predetermined threshold (e.g., 99.99%), guiding the decision to accept or reject the hypothesis and thereby determining whether the suspect is a contributor to the mixed sample.
Results and discussion
CE-STR profiles analysis
The CE-STR profiles analysis revealed no specific PCR amplification products detected in Sects. 1–1, 1–2, 1–3, 2–2, and 8−2, with typical profiles illustrated in Fig. 2a. Sections 2–3 and 4−1 exhibited mixed genotypes, as shown in Fig. 2b, while the STR profiles detected in the remaining sections matched with the victim’s genotypes. Using GMP 3 software with the mixed STR profiles as input and the victim’s STR profiles as a reference, the weights of autosomal STR loci genotypes did not yield convincing results for the contributor other than the victim (Table 1). Consequently, the suspect’s STR genotypes could not be precisely inferred, preventing confirmation of the suspect’s STR presence in the DNA mixtures.
Given that the CE-STR profiles indicated degradation and extreme mixing ratios in some DNA samples, MNPs were used for further analysis. Sections 2–3, 3−2, 4−1, and the suspect’s DNA sample were analyzed, revealing single-source MNP genotypes for Sect. 3−2 and the suspect, with most loci exhibiting no more than two alleles. Sections 2–3 and 4−1 showed mixed MNP genotypes, consistent with the CE-STR results. The number of contributors in Sects. 2–3 and 4−1 was determined to be two based on the polynomial distribution of alleles and the maximum likelihood function.
Suspect DNA presence probability calculation
The probability of the suspect’s DNA presence in the mixed samples was calculated. In Sect. 4−1, at least one allele of the suspect matched the alleles at each locus. However, in Sect. 2–3, some of the suspect’s alleles were absent at certain loci, resulting in a nearly zero match probability, effectively ruling out the suspect’s DNA presence in Sect. 2–3. On the other hand, the presence of the suspect’s MNP genotypes in Sect. 4−1 was confirmed with a probability of 1–8.41 × 10− 6.
Anzeige
The location of Sect. 4−1 was on the lowest wooden bar of the campstool (Fig. 1), and this evidence contradicted the suspect’s statement that he only touched the upper part of the campstool when the victim handed it to him, aiding the police in resolving the case.
Comparison of CE-STR and MNP analysis
Current forensic researches on DNA mixtures focus on optimizing biological technology systems, selecting genetic markers, and interpreting genotype data. While CE-STR systems and emerging NGS-STR systems are being optimized, issues such as the stutter sequence and amplification imbalances in STR amplification persist [16, 17]. Additionally, the longer length of STR profiles compared to SNPs can result in incomplete profiles for minor contributor’s DNA below 20% [18‐20], limiting STR application in low template DNA identification. Although new genetic markers like DIP-STRs [21] and SNP-STRs [22] are effective for extremely unbalanced DNA mixtures, they are less effective for complex mixtures with more than two contributors. MHs have shown promise for such analyses, and MNPs, being smaller in molecular size and greater in number [23], offer advantages in identifying degraded samples and complex DNA mixtures.
Several software tools have been developed for mixed DNA profile deconvolution [24]. In this case, GPM 3 software was used for STR profiles [14], incorporating a fully continuous model and considering biological parameters like peak height, pre-stutter ratio, and fragment sizes. However, validation studies have shown that likelihood ratios (LRs) for individual mixture contribution, especially minor ones, are lower in low-amount DNA samples [14], explaining our initial failure in interpreting the mixed STR profiles.
In contrast, MNP analysis involves calculating the presence probability of the suspect’s DNA rather than direct deconvolution, ensuring an accurate assessment of the evidence’s weight and admissibility in court. The high-throughput sequencing capabilities and shorter fragment lengths of MNPs allowed for effective detection of low-amount or degraded DNA, yielding comprehensive genotype data for probability calculations. These findings underscore the advantages of MNPs over STRs in forensic DNA analysis.
Anzeige
In conclusion, multi-SNPs are valuable genetic markers for identifying DNA mixtures, particularly those that are trace in amount and degraded. This case demonstrates that MNPs, combined with probability calculations, can provide robust evidence for DNA mixture contributors. Future developments in new algorithms may enable direct inference of contributor MNP genotypes in DNA mixtures.
Fig. 1
The diagram of the sampling areas of the campstool
Fig. 2
CE-STR profiles of DNA samples from the campstool sections. (a) Typical profiles showing no specific PCR amplification products detected in Sect. 2–2. (b) Mixed genotypes observed in Sect. 4−1
Table 1
The typical deconvolution results of Sect. 4−1 by GPM 3
Marker | Genotype | Weight | Component > 90% |
|---|---|---|---|
D8S1179 | 11, 15 | 40.8923% | |
13, 15 | 29.2613% | ||
D21S11 | 29, 30 | 47.7912% | |
29, 29 | 43.7586% | ||
D7S820 | 11, 11 | 41.4834% | |
8, 11 | 38.6519% | ||
CSF1PO | 12, 12 | 41.2676% | |
11, 12 | 36.2056% | ||
D3S1358 | 15, 16 | 61.4539% | |
15, 17 | 12.0282% | ||
TH01 | 6, 7 | 59.2706% | |
7, 7 | 15.2560% | ||
D13S317 | 8, 11 | 29.7037% | |
8, 12 | 25.1534% | ||
D16S539 | 11, 11 | 72.0477% | |
11, -1 | 22.0897% | ||
D19S433 | 12, 15 | 37.2809% | |
13, 15 | 31.4952% | ||
vWA | 17, 20 | 70.3016% | |
16, 17 | 7.8010% | ||
TPOX | 8, 8 | 89.4720% | |
8, -1 | 9.5020% | ||
AM | X, Y | 97.0743% | X, Y |
X, X | 2.9257% | ||
D5S818 | 12, 13 | 85.9430% | |
13, 13 | 7.0430% |
Anzeige
Acknowledgements
We thank Professor Penghai’s team from Jianghan University for their assistance with the MNP calculations in this work.
Declarations
Ethics approval and consent to participate
The study was conducted according to the guidelines of the Declaration of Helsinki, and was approved by the Ethics Committee of Academy of Forensic Science.
Consent for publication
The campstool, the victim’s STR genotypes, and the suspect’s blood sample were provided by the policeman. Informed consent was obtained for the publication of the data and findings from this study.
Research involving human participants and/or animals
Human participants.
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
