Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Oct 20;12(11):1649.
doi: 10.3390/genes12111649.

Precision DNA Mixture Interpretation with Single-Cell Profiling

Jianye Ge  1   2 , Jonathan L King  1 , Amy Smuts  1 , Bruce Budowle  1   2
Affiliations

Precision DNA Mixture Interpretation with Single-Cell Profiling

Jianye Ge et al. Genes (Basel). .

Abstract

Wet-lab based studies have exploited emerging single-cell technologies to address the challenges of interpreting forensic mixture evidence. However, little effort has been dedicated to developing a systematic approach to interpreting the single-cell profiles derived from the mixtures. This study is the first attempt to develop a comprehensive interpretation workflow in which single-cell profiles from mixtures are interpreted individually and holistically. In this approach, the genotypes from each cell are assessed, the number of contributors (NOC) of the single-cell profiles is estimated, followed by developing a consensus profile of each contributor, and finally the consensus profile(s) can be used for a DNA database search or comparing with known profiles to determine their potential sources. The potential of this single-cell interpretation workflow was assessed by simulation with various mixture scenarios and empirical allele drop-out and drop-in rates, the accuracies of estimating the NOC, the accuracies of recovering the true alleles by consensus, and the capabilities of deconvolving mixtures with related contributors. The results support that the single-cell based mixture interpretation can provide a precision that cannot beachieved with current standard CE-STR analyses. A new paradigm for mixture interpretation is available to enhance the interpretation of forensic genetic casework.

Keywords: DNA forensics; DNA mixture; clustering algorithm; consensus profile; mixture interpretation; number of contributors; single-cell.

PubMed Disclaimer

Conflict of interest statement

A patent application in pending.

Figures

Figure 1
Figure 1
The accuracies of NOC estimation for 3-person mixtures with 2, 6, and 12 single cells with various ADO and ADI rates in three scenarios: (a) Diploid cells with three unrelated contributors, (b) Diploid cells with three family trio contributors (father, mother, and child), and (c) Haploid cells with three unrelated contributors. D = 0.2; e = 0.01; 10,000 simulations for each mixture scenario.
Figure 1
Figure 1
The accuracies of NOC estimation for 3-person mixtures with 2, 6, and 12 single cells with various ADO and ADI rates in three scenarios: (a) Diploid cells with three unrelated contributors, (b) Diploid cells with three family trio contributors (father, mother, and child), and (c) Haploid cells with three unrelated contributors. D = 0.2; e = 0.01; 10,000 simulations for each mixture scenario.
Figure 2
Figure 2
Mixture visualization examples with 3D MDS plots: (a) A family trio diploid cell mixture (53, 53, 53) incorrectly clustered with the EM algorithm and Silhouettes method as two contributors. (b) A family trio diploid cell mixture (8,24,48) correctly clustered with the EM algorithm and Silhouettes method as three contributors. (c) A 3-person haploid cell mixture (2,6,12) incorrectly clustered with the EM algorithm and Silhouettes method as two contributors. (d) A 3-person diploid cell mixture (4, 4, 12) incorrectly clustered with the EM algorithm and Silhouettes method as four contributors. Each color represents each cluster. Rotatable plots of these figures can be found in Supplementary Figure S1.
Figure 2
Figure 2
Mixture visualization examples with 3D MDS plots: (a) A family trio diploid cell mixture (53, 53, 53) incorrectly clustered with the EM algorithm and Silhouettes method as two contributors. (b) A family trio diploid cell mixture (8,24,48) correctly clustered with the EM algorithm and Silhouettes method as three contributors. (c) A 3-person haploid cell mixture (2,6,12) incorrectly clustered with the EM algorithm and Silhouettes method as two contributors. (d) A 3-person diploid cell mixture (4, 4, 12) incorrectly clustered with the EM algorithm and Silhouettes method as four contributors. Each color represents each cluster. Rotatable plots of these figures can be found in Supplementary Figure S1.
Figure 3
Figure 3
IBS distributions of unrelated and related pairs (UR, PC, and FS, in dotted lines) and pairs of the true profile versus the consensus profile with a given number of cells (in solid lines) based on (a) diploid cells and (b) haploid cells. D = 0.2; e = 0.01; 1,000,000 simulations for each distribution. The IBS of “n cell(s)” is the IBS between true genotypes and the consensus genotypes from n cell(s). These distributions are based on 21 GlobalFiler markers and thus the maximum IBS is 42. The consensus profile might contain missing alleles, and these missing alleles were excluded in counting IBS.
Figure 4
Figure 4
Consensus accuracy as a function of the number of cells in the cluster, with 1:1 ratio mixtures. D = 0.2; e = 0.01; 10,000 simulations for each mixture. The bars represent the 95% confidence intervals.
Figure 5
Figure 5
IBS distributions of unrelated and related pairs (UR, PC, and FS, in dotted lines) and pairs of the true profile versus the consensus profile with a given number of cells (in solid lines) for (a) 1 diploid cell and (b) 4 haploid cells. 1,000,000 simulations for each distribution. Missing alleles are excluded in counting IBS.
Figure 6
Figure 6
The accuracies of consensus for 3-person mixtures with 2, 6, and 12 single cells with various ADO and ADI rates in three scenarios: (a) diploid cells with three unrelated contributors, (b) diploid cells with three family trio contributors (father, mother, and child), and (c) haploid cells with three unrelated contributors.
Figure 6
Figure 6
The accuracies of consensus for 3-person mixtures with 2, 6, and 12 single cells with various ADO and ADI rates in three scenarios: (a) diploid cells with three unrelated contributors, (b) diploid cells with three family trio contributors (father, mother, and child), and (c) haploid cells with three unrelated contributors.
Figure 7
Figure 7
The interpretation workflow of single-cell profiles generated from a mixture.
See this image and copyright information in PMC

References

    1. Bright J.-A., Taylor D., McGovern C., Cooper S., Russell L., Abarno D., Buckleton J. Developmental validation of STRmix™, expert software for the interpretation of forensic DNA profiles. Forensic Sci. Int. Genet. 2016;23:226–239. doi: 10.1016/j.fsigen.2016.05.007. - DOI - PubMed
    1. Gill P., Haned H., Eduardoff M., Santos C., Phillips C., Parson W. The open-source software LRmix can be used to analyse SNP mixtures. Forensic Sci. Int. Genet. Suppl. Ser. 2015;5:e50–e51. doi: 10.1016/j.fsigss.2015.09.020. - DOI - PubMed
    1. Perlin M.W., Legler M.M., Spencer C.E., Smith J.L., Allan W.P., Belrose J.L., Duceman B.W. Validating TrueAllele DNA mixture interpretation. J. Forensic Sci. 2011;56:1430–1447. doi: 10.1111/j.1556-4029.2011.01859.x. - DOI - PubMed
    1. Gill P., Jeffreys A.J., Werrett D.J. Forensic application of DNA ‘fingerprints’. Nature. 1985;318:577–579. doi: 10.1038/318577a0. - DOI - PubMed
    1. Voorhees J.C., Ferrance J.P., Landers J.P. Enhanced elution of sperm from cotton swabs via. enzymatic digestion for rape kit analysis. J. Forensic Sci. 2006;51:574–579. doi: 10.1111/j.1556-4029.2006.00112.x. - DOI - PubMed

Publication types

LinkOut - more resources