Second GHEP-ISFG exercise for DVI: "DNA-led" victims' identification in a simulated air crash.

The Spanish and Portuguese-Speaking Working Group of the International Society for Forensic Genetics (GHEP-ISFG) has organized a second collaborative exercise on a simulated case of Disaster Victim Identification (DVI), with the participation of eighteen laboratories. The exercise focused on the analysis of a simulated plane crash case of medium-size resulting in 66 victims with varying degrees of fragmentation of the bodies (with commingled remains). As an additional difficulty, this second exercise included 21 related victims belonging to 6 families among the 66 missings to be identified. A total number of 228 post-mortem samples were represented with aSTR and mtDNA profiles, with a proportion of partial aSTR profiles simulating charred remains. To perform the exercise, participants were provided with aSTR and mtDNA data of 51 reference pedigrees -some of which deficient-including 128 donors for identification purposes. The exercise consisted firstly in the comparison of the post-mortem genetic profiles in order to re-associate fragmented remains to the same individual and secondly in the identification of the re-associated remains by comparing aSTR and mtDNA profiles with reference pedigrees using pre-established thresholds to report a positive identification. Regarding the results of the post-mortem samples re-associations, only a small number of discrepancies among participants were detected, all of which were from just a few labs. However, in the identification process by kinship analysis with family references, there were more discrepancies in comparison to the correct results. The identification results of single victims yielded fewer problems than the identification of multiple related victims within the same family groups. Several reasons for the discrepant results were detected: a) the identity/non-identity hypotheses were sometimes wrongly expressed in the likelihood ratio calculations, b) some laboratories failed to use all family references to report the DNA match, c) In families with several related victims, some laboratories firstly identified some victims and then unnecessarily used their genetic information to identify the remaining victims within the family, d) some laboratories did not correctly use "prior odds" values for the Bayesian treatment of the episode for both post-mortem/post-mortem re-associations as well as the ante-mortem/post-mortem comparisons to evaluate the probability of identity. For some of the above reasons, certain laboratories failed to identify some victims. This simulated "DNA-led" identification exercise may help forensic genetic laboratories to gain experience and expertize for DVI or MPI in using genetic data and comparing their own results with the ones in this collaborative exercise.

Results of the 1999-2000 collaborative exercise and proficiency testing program on mitochondrial DNA of the GEP-ISFG: an inter-laboratory study of the observed variability in the heteroplasmy level of hair from the same donor.

The Spanish and Portuguese working group (GEP) of international society for forensic genetics (ISFG) 1999-2000 collaborative exercise on mitochondrial DNA (mtDNA) included the analysis of four bloodstain samples and one hair shaft sample by 19 participating laboratories from Spain, Portugal and several Latin-American countries. A wide range of sequence results at position 16,093 of the HV1 (from T or C homoplasmy to different levels of heteroplasmy) were submitted by the different participating laboratories from the hair shaft sample during the first phase of this exercise. During the discussion of these results in the Annual GEP-ISFG 2000 Conference a second phase of this exercise was established with two main objectives: (i) to evaluate the incidence of the HV1 sequence heteroplasmy detected in Phase I across different sample types from the same donor including blood, saliva, and hair shafts, (ii) to perform a technical review of the electropherograms to evaluate the relative levels of heteroplasmies obtained by the different laboratories and also to examine the source of possible errors detected in Phase I. Anonymous review of the raw sequence data permitted the detection of three transcription errors and three errors due to methodological problems. Highly variable levels of heteroplasmy were found in the hair shaft and more stability in blood and saliva. Three laboratories found variable levels of heteroplasmy at position 16,093 across adjacent fragments from the same hair shaft. Two laboratories also described more than one heteroplasmic position from a single hair. The relevance of these findings for the interpretation of mtDNA data in the forensic context is also discussed.