Next generation sequencing technology in Second World War victim identification

Skeletal remains are among the most challenging biological samples for successful STR typing due to the degradation, presence of inhibitors and possible contamination []. In recent years the recovery and analysis of DNA from skeletal remains has been applied to several contexts ranging from disaster victim identification to the identification of missing persons and the victims of war conflicts []. DNA analyses of WWII skeletons have been used for identifying victims from mass graves found in Poland [], Bosnia and Herzegovina [], Slovenia [], Russia [] and others. The Commission of the Government of the Republic of Slovenia on Concealed Mass Graves has registered more than 600 WWII hidden mass graves in the last 25 years. The aim of the study presented was to perform the genetic identification of Slovenian victims killed in 1942 and buried in a hidden mass grave in Babna Gora (7 victims were missing). In 2015, three incomplete female skeletons were excavated from the shallow grave and we have been asked for genetic identification by comparison with two living relatives.

From badly preserved incomplete skeletal remains, genetic typing was successful only for skeleton 2 and the results for that skeleton are presented. Femur, tibia and the left second molar were analysed and compared presumptive nephew and niece. We cleaned the bones and tooth, removed surface contamination and ground them into powder. Prior to DNA isolation 0.5 g of powder was decalcified using EDTA. DNA was purified in the Biorobot EZ1 device (Qiagen) using the EZ1 DNA Investigator Kit (Qiagen). Genomic DNA extraction was performed twice from each sample. The nuclear DNA of the samples was quantified with the PowerQuant kit (Promega) in an ABI PRISM 7500 Real Time PCR System (AB) and STR typing performed using ESI 17 (Promega) and NGM (TFS) STR amplification kits. The PCR products were separated on an ABI 3130 Genetic Analyzer (AB). The profiles were determined using the Data Collection v 3.0 and GeneMapper ID v 3.2 (AB) software. The Precision ID GlobalFiler NGS STR Panel (TFS) was used on NGS Ion S5 System (TFS) to increase the number of loci analysed. DNA library preparations was run automatically on the Ion Chef System (TFS) using the Precision ID DL8 Kit for Chef, for which 15 μl of DNA sample was used. The concentration of the library pool was determined in duplicate by qPCR with the Ion Library TaqMan Quantitation Kit (TFS) using 7500 Real-Time PCR with HID Real-Time PCR Analysis Software version 1.2 in line with the manufacturer’s user guide (GlobalFiler NGS STR panel Manual 2017). The library pool diluted to 50 pM was used for fully automated DNA template preparation on the Ion Chef System. The GlobalFiler NGS STR panel template was prepared using the Ion S5 Precision ID Chef Reagents and loaded on an Ion 520 sequencing chip following the manufacturer’s guidelines (GlobalFiler NGS STR panel Manual 2017). The calculation of likelihood ratios (LR) and PP was performed with the DNA VIEW software (C. Brenner) and the Converge Software - Kinship Analysis (TFS). The prior probability of 0.125 was set based on the number of victims reported and a recommended PP (for kinship) of 99.9% was used with the goal of high confidence of correct identification of victim in the mass grave [, , ]. For traceability in the event of contamination, we created an elimination database including profiles of all persons that had been in contact with the excavated skeletal remains.

Up to 0.79 ng DNA/g of powder was obtained from three skeletal elements analysed from skeleton 2 and consensus profile (determined from femur, tibia and left second molar) was compared with family references. LR was calculated for autosomal STRs only, because no appropriate family references were available for the comparison of mtDNA haplotypes (for identification of the aunt only nephew and niece on paternal line were traced after more than 70 years) and too low PP was obtained after ESI 17 and NGM typing. Since PP of 99.9% was followed [, , ], the Precision ID GlobalFiler NGS STR Panel (TFS) was used and after the analysis of additional STR loci, the statistical calculation showed the PP of 99.99986% indicating that a sufficient number of genetic markers were investigated in identifying skeletal remains of the aunt. When comparing the genetic profiles of femur, tibia and the left second molar with an elimination database (profiles of individuals that participated in the excavation, anthropological investigation and genetic analyses) no match was found.

For extracting genomic DNA from the Babna Gora mass grave tooth and bones, a highly efficient extraction protocol that comprises generation of fine powder, complete demineralisation, and efficient purification was used [] and successful autosomal STR typing using the Precision ID GlobalFiler NGS STR Panel (TFS) allowed the identification of the aunt with sufficient statistical support and with high confidence of correct identification (Fig. 1).

In presented case, NGS technology proved to be a powerful tool for increasing the number of autosomal STRs needed for identification of WWII victims when linear markers cannot be used for comparison and only distant relatives are available for analyses. After more than 70 years, the skeletal remains were returned to the living relatives and buried in a family tomb.

The authors would like to thank the Slovenian Government Commission on Concealed Mass Graves for its support in the exhumations of WWII victims. This study was financially supported by the Slovenian Research Agency (the project “Determination of the most appropriate skeletal elements for molecular genetic identification of aged human remains” (J3-8214)). The research project was approved by the Slovenian Medical Ethics Committee (KME118/05/15) and informed agreements were obtained from living relatives and individuals for inclusion in the elimination database.