A hybrid approach to increase the informedness of CE-based data using locus-specific thresholding and machine learning

Marciano, Michael A.; Williamson, Victoria R.; Adelman, Jonathan D.

doi:10.1016/j.fsigen.2018.03.017

Cited by 11 publications

(17 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The cycle set average peak heights increased as the cycle number increased with 29 A more detailed, locus-speci c, analysis of average peak heights shows the same trend, where average peak height (PCR product) increases within each locus as cycle number increases (Table 1, Figure 1 and Additional le 1- Figure 1S). The overall mean peak height derived from the locus-speci c average peak heights resulted in mean of 81.4 ± 42.8, 164.9 ± 100.3 and 253.8 ± 162.6 RFU for 29, 30 and 31 cycles respectively.…”

Section: Peak Height Evaluationmentioning

confidence: 73%

“…These values were chosen due to widespread use of analytical thresholds between 100 and 150RFU; the 10 RFU threshold is used to demonstrate the maximum information gain. It is not uncommon that instruments would yield a threshold of between 15 and 20RFU when calculating analytical thresholds using mean + 3 standard deviations (a 99% con dence interval) [29]. Average peak heights at each autosomal locus was calculated as described in [30], where the peak heights of heterozygotic loci were averaged and the peak heights of homozygous loci were divided by two.…”

Section: Sample Processing and Analysismentioning

confidence: 99%

See 1 more Smart Citation

Revisiting Single Cell Analysis in Forensic Science

Watkins

Myers²,

Xavier

et al. 2020

Preprint

View full text Add to dashboard Cite

Background Forensic science has yet to take full advantage of single cell analysis. Its greatest benefit is the ability to alleviate the challenges associated with DNA mixture analysis, which, despite the emergence of probabilistic genotyping, remains a significant hurdle in forensic science. Many of the factors that cause complexity in mixture interpretation are absent in single cell analyses — multiple contributors and varied levels of contribution and allele masking. Recent technological innovations have been developed within the scientific community that mitigate the historical risks of single cell analysis in forensic casework. This study revisits single cell analyses in the context of forensic identification, introducing previously unseen depth to the characterization of data generated from single cells using a novel pipeline that includes recovery of single cells using the DEPArray™ NxT and amplification using the PowerPlex Fusion 6c kit with varied PCR cycles (29, 30, and 31); the resulting allelic signal was assessed using analytical thresholds of 10, 100 and 150RFU. Results The mean peak heights across the sample sets generally increased as cycle number increased, 75.0 ± 85.3, 147.1 ± 172.6, and 226.1 ± 298.2 RFU, for 29, 30, and 31 cycles, respectively. The average proportion of allele/locus dropout was most significantly impacted by changes in the detection threshold, whereas increases f PCR cycle number had less impact. Overall data quality improved notably when increasing PCR from 29 to 30 cycles; however, less improvement and more volatility introduced at 31 cycles. The average random match probabilities (RMP) for the 29, 30, and 31 cycle sets at 150RFU are 1 in 2.4 × 1018 ± 1.46 × 1019, 1 in 1.49 × 1025 ± 5.8 × 1025, and 1 in 1.83 × 1024 ± 8.09 × 1024, respectively. Conclusions This study introduces and optimizes an analytical pipeline to generate ample high quality data from single cells that can be used in forensic analyses, thus removing the need for complex mixture interpretation. An analysis of the impact of PCR cycles on single cells revealed a significant overall improvement in the amount and quality of interpretable data when increasing from 29 to 30 cycles, while no such improvement was noted when increasing to 31 cycles.

show abstract

Section: Peak Height Evaluationmentioning

confidence: 73%

Section: Sample Processing and Analysismentioning

confidence: 99%

Revisiting Single Cell Analysis in Forensic Science

Watkins

Myers²,

Xavier

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The use of different analytical thresholds also predictably led to decreased trimming accuracy. The use of a dynamic, locus‐ and sample‐specific analytical threshold is used to minimize the number of peaks that are incorrectly trimmed, regardless of the number of artifacts that have not been trimmed, with the underlying assumption that subsequent modeling will more precisely target artifacts such as pull‐up and stutter . The dynamic threshold used in this study trimmed far fewer peaks, in general, than would a static threshold, and left a larger number of pull‐up artifacts for the pull‐up models to detect and remove, thus reflecting the higher trimming accuracy than what was observed for 50 RFU and 150 RFU thresholds.…”

Section: Discussionmentioning

confidence: 99%

“…The resulting data, including trace data – the x‐y data associated with the entirety of the electropherogram – were exported from Armed Expert. Analytical thresholds were calculated using either a dynamic, locus‐ and sample‐specific method or static thresholds at 50 and 150 RFU. The data set was filtered to include only those artifactual peaks caused by incomplete removal of spectral overlap, i.e., non‐allelic peaks within 0.6 bases of allelic peaks in separate dye channels.…”

Section: Methodsmentioning

confidence: 99%

“…For each DNA profile in the data set, three analytical thresholds were applied in separate analyses for each locus: a dynamic, locus‐ and sample‐specific threshold (LSST) , a 50 RFU static threshold, and a 150 RFU static threshold. The models used for all analyses were optimized using LSST to enable prediction of pull‐up even at very low peak heights.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Automated detection and removal of capillary electrophoresis artifacts due to spectral overlap

et al. 2019

Self Cite

View full text Add to dashboard Cite

While DNA detection using capillary electrophoresis has enabled improvements in both resolution and throughput, the use of CE – particularly with multiple dye channels – can introduce artifacts that can complicate analyses. Undetected pull‐up artifacts can pose a challenge to investigators, especially in low‐level samples, while partial pull‐up peaks can distort peak height balance within a locus and impact the downstream likelihood ratio. Current methods for addressing pull‐up are typically manually implemented. This study presents an effective alternative: a series of mathematical models, created using symbolic regression achieved through genetic programming. The models estimate the amount of pull‐up expected in a peak from a true allele for a given dye‐dye relationship and instrument type. This leads to the removal of artifactual pull‐up peaks and peak height corrections when pull‐up is present within true alleles. When models are used in conjunction with a dynamic threshold, pull‐up peaks were automatically detected and removed with an accuracy rate of 96.1%. The removal of partial pull‐up from true allele peaks led to a more accurate heterozygote balance for the affected locus. These models have been optimized for use with any analytical threshold and can be implemented by any lab using a 3100 or 3500 instrument series.

show abstract

Forensic ancestry analysis in two Chinese minority populations using massively parallel sequencing of 165 ancestry‐informative SNPs

Wang

et al. 2018

Electrophoresis

View full text Add to dashboard Cite

Ancestry inference based on SNPs with marked allele frequency differences in diverse populations (called ancestry-informative SNP (AISNP)) is rapidly developed with the technology advancements of massively parallel sequencing. Despite the decade of exploration and broad public interest in the peopling of East-Asians, the genetic landscape of Chinese Silk Road populations based on the AISNPs is still little known. In this work, 206 unrelated individuals from Chinese Uyghur and Hui populations were firstly genotyped by 165 AISNPs (The Precision ID Ancestry Panel) using the Ion Torrent Personal Genome Machine system. The ethnic origin of two investigated populations and population structures and genetic relationships were subsequently investigated. The 165 AISNPs panel not only can differentiate Uyghur and Hui populations but also has potential applications in individual identification. Comprehensive population comparisons and admixture estimates demonstrated a predominantly higher European-related ancestry (36.30%) in Uyghurs than Huis (3.66%). Overall, the Precision ID Ancestry Panel can provide good resolution at the intercontinental level, but has limitations on the genetic homogeneous populations, such as the Hui and Han. Additional population-specific AISNPs remain necessary to get better-scale resolution within geographically proximate populations in East Asia.

show abstract

A hybrid approach to increase the informedness of CE-based data using locus-specific thresholding and machine learning

Cited by 11 publications

References 29 publications

Revisiting Single Cell Analysis in Forensic Science

Revisiting Single Cell Analysis in Forensic Science

Automated detection and removal of capillary electrophoresis artifacts due to spectral overlap

Forensic ancestry analysis in two Chinese minority populations using massively parallel sequencing of 165 ancestry‐informative SNPs

Contact Info

Product

Resources

About