DNA Identification in Mass Fatality Incidents
Amanda Sozer
We found that instituting quality checks throughout the identification process ultimately saved time and effort. By continually validating the accuracy of the data and results at each step in the analysis, we could identify potential issues before they became impediments to an identification.
Software is not only a case-tracking tool. It is also a critical component of a DNA laboratory’s quality assurance and quality control programs.
Quality metrics collected and tracked through software are used to refine and improve the laboratory’s quality assurance plan, and software tools often are employed as quality control mechanisms. Mass fatality incident responses have several, specific quality control needs:
- Identify conflicting STR results. Remains samples and personal items may not yield usable DNA profiles on the first analysis attempt. The laboratory may choose to reanalyze these samples under altered conditions in the hope of producing a complete—or a more complete—profile. The laboratory will need to compare the results from each analysis to identify and resolve conflicts.
- Identify conflicting results from different DNA technologies. When multiple DNA technologies are used, the laboratory will need to review previously reported identifications to ensure that results from the new technologies are consistent. For example, a remains sample and a personal item may match with STRs but not with mitochondrial DNA (mtDNA).
- Identify fortuitous matches. Partial profiles resulting from sample degradation are a common occurrence in mass fatality incidents. A partial profile may match several reference samples fortuitously, particularly if the matching algorithm allows for the possibility of allelic dropout. The DNA analyst must review all of the candidate matches and choose an appropriate course of action. The software should produce a work list that allows the DNA analyst to record free-text comments about each discrepancy.
If the laboratory chooses to outsource samples to partner laboratories, these additional quality control tools should be considered:
- Data file validation. The managing laboratory may want to validate the format and content of data files that are provided by partner laboratories. Fields that may be validated include sample names (to ensure the appropriate naming scheme was applied) and loci and allele values.
- Blind-control verification. One method of monitoring quality in partner laboratories is to institute a blind-control program (see chapter 14, Quality Control). To partner laboratories, blind controls appear to be normal samples; however, their profiles have already been determined by the managing laboratory. The managing laboratory randomly places blind controls into the batches of samples (or microtiter plates) that are shipped to partner labs. The blind controls usually are renamed so that they are indistinguishable to the partner laboratories from normal samples. Then, the managing laboratory checks the data files that are produced by partner laboratories for blind controls and verifies them against the known DNA profiles.
