Inter-laboratory study on standardized MPS libraries: evaluation of performance, concordance, and sensitivity using mixtures and degraded DNA.

We present results from an inter-laboratory massively parallel sequencing (MPS) study in the framework of the SeqForSTRs project to evaluate forensically relevant parameters, such as performance, concordance, and sensitivity, using a standardized sequencing library including reference material, mixtures, and ancient DNA samples. The standardized library was prepared using the ForenSeq DNA Signature Prep Kit (primer mix A). The library was shared between eight European laboratories located in Austria, France, Germany, The Netherlands, and Sweden to perform MPS on their particular MiSeq FGx sequencers. Despite variation in performance between sequencing runs, all laboratories obtained quality metrics that fell within the manufacturer's recommended ranges. Furthermore, differences in locus coverage did not inevitably adversely affect heterozygous balance. Inter-laboratory concordance showed 100% concordant genotypes for the included autosomal and Y-STRs, and still, X-STR concordance exceeded 83%. The exclusive reasons for X-STR discordances were drop-outs at DXS10103. Sensitivity experiments demonstrated that correct allele calling varied between sequencing instruments in particular for lower DNA amounts (≤ 125 pg). The analysis of compromised DNA samples showed the drop-out of one sample (FA10013B01A) while for the remaining three degraded DNA samples MPS was able to successfully type ≥ 87% of all aSTRs, ≥ 78% of all Y-STRs, ≥ 68% of all X-STRs, and ≥ 92% of all iSNPs demonstrating that MPS is a promising tool for human identity testing, which in return, has to undergo rigorous in-house validation before it can be implemented into forensic routine casework.

Evaluation of the VISAGE basic tool for appearance and ancestry inference using ForenSeq® chemistry on the MiSeq FGx® system.

The possibility of providing investigative leads when conventional DNA identification methods fail to solve a case can be of extreme relevance to law enforcement. Therefore, the forensic genetics community has focused research towards the broadened use of DNA, particularly for prediction of appearance traits, bio-geographical ancestry and age. The VISible Attributes through GEnomics (VISAGE) Consortium expanded the use of DNA phenotyping by developing new molecular and statistical tools for appearance, age and ancestry prediction. The VISAGE basic tool for appearance (EVC) and ancestry (BGA) prediction was initially developed using Ampliseq chemistry, but here is being evaluated using ForenSeq chemistry. The VISAGE basic tool offers a total of 41 EVC and 115 BGA SNPs and thus provides more predictions, i.e., skin color, than achieved with the ForenSeq DNA Signature Prep kit that is based on 24 EVC and 56 BGA SNPs. Five VISAGE laboratories participated in collaborative experiments to provide foreground for developmental validation of the assay. Assessment of assay performance and quality metrics, reproducibility, sensitivity, inhibitor tolerance and species specificity are described. Furthermore, the assay was tested using challenging samples such as mock casework samples and artificially degraded DNA. Two different analysis strategies were applied for this study and output on genotype calls and read depth was compared. Overall, inter-laboratory, inter-method and concordance with publicly available data were analysed and compared. Finally, the results showed a reliable and robust tool, which can be easily applied for laboratories already using a MiSeq FGx with ForenSeq reagents.

Recommendations of the DNA Commission of the International Society for Forensic Genetics (ISFG) on quality control of autosomal Short Tandem Repeat allele frequency databasing (STRidER).

The statistical evaluation of autosomal Short Tandem Repeat (STR) genotypes is based on allele frequencies. These are empirically determined from sets of randomly selected human samples, compiled into STR databases that have been established in the course of population genetic studies. There is currently no agreed procedure of performing quality control of STR allele frequency databases, and the reliability and accuracy of the data are largely based on the responsibility of the individual contributing research groups. It has been demonstrated with databases of haploid markers (EMPOP for mitochondrial mtDNA, and YHRD for Y-chromosomal loci) that centralized quality control and data curation is essential to minimize error. The concepts employed for quality control involve software-aided likelihood-of-genotype, phylogenetic, and population genetic checks that allow the researchers to compare novel data to established datasets and, thus, maintain the high quality required in forensic genetics. Here, we present STRidER (http://strider.online), a publicly available, centrally curated online allele frequency database and quality control platform for autosomal STRs. STRidER expands on the previously established ENFSI DNA WG STRbASE and applies standard concepts established for haploid and autosomal markers as well as novel tools to reduce error and increase the quality of autosomal STR data. The platform constitutes a significant improvement and innovation for the scientific community, offering autosomal STR data quality control and reliable STR genotype estimates.

Reduction of PCR-amplifiable DNA by ethylene oxide treatment of forensic consumables.

A reliable method to provide molecular biology products free of contaminating DNA is of forensic interest. Ethylene oxide (EO) treatment has been demonstrated as an effective method in published studies. This study aimed to address some additional experiments that are closer to forensic practice. In the first part of this study, different consumables such as cotton swabs, latex gloves and micro test tubes were spiked with saliva, blood and skin cells to mimic a real-life contamination scenario. EO treatment was performed for a period of 3, 5, 7, and 10h, respectively. For comparison, gamma and electron beam treatment was applied. In the second part of this study, a cell culture line (K562) was used to apply defined cell counts on cotton swabs followed by EO treatment for 3 and 5h. After extraction of samples, the DNA content was quantified using a real-time PCR based system. STR analysis was performed using a latest generation STR kit to meet current sensitivity limits. A good correlation of real-time PCR results and STR results was observed. This work confirmed the findings of earlier studies showing that chemical EO treatment is much more successful in reducing the amount of PCR-amplifiable DNA than ionising radiation. Furthermore, the efficacy of EO treatment is affected by the nature of the samples. DNA in saliva was more susceptible to damage by EO gas than DNA in blood. Our results show, that accessibility of the sample to EO gas has a strong influence on the method's efficiency. While treatment of samples on cotton swabs packed into gas-permeable bags was very successful, samples inside a closed micro test tube were resistant to the same treatment conditions. Our work with defined K562 cell numbers and multi-copy quantitative PCR could show that a 5h EO treatment results in a 10(5) fold reduction of PCR-amplifiable DNA. Corresponding STR-PCR results also show only sporadic allele calls in the Mini-loci range, providing a reliable interpretation of forensic analysis. Finally, we do recommend an EO treatment of forensic consumables and a multi-copy quantitative PCR approach to establish reliable treatment conditions.