Rapid DNA technology usage for forensic-type samples has been gaining momentum amid the manufacturer modifications to the originally commercialized cartridges.   While the technology is not as sensitive as conventional laboratory methods, it can prove beneficial in certain scenarios including disaster victim identification.  This study aimed to (1) determine if probabilistic genotyping – specifically STRMix v 2.12 – can be used in conjunction with RapidHIT ID Intel Plus generated data; (2) establish the parameters and limitations of usage; and (3) compare it to the results achieved when processing the same data using conventional means.  
Based on ground truth and intuitive alignment with expert interpretations, these studies demonstrate that STRmix v2.12 can be used for the interpretation of evidentiary DNA typing results that originate from 1-3 donors.  This assertion is based on information gathered through studies which included samples comprised of various mixture proportions and the interpretation of data assuming N+/- 1 donor using 3p as the highest number of contributors to a given mixture. 

Due to the oftentimes ‘messy’ baseline of rapid data, many additional artifacts and stutter peaks are detected when compared to traditional analysis; this requires greater analyst time evaluating electropherograms prior to STRmix analysis.  Some peaks are often unresolved from neighboring alleles and artifacts (i.e., 1 base pair resolution issues).  Additionally, missing stutter flags were frequently observed for stutter types (e.g., TH01 -2 bp stutter) that were detected during our stutter analysis.  Many of the unintuitive diagnostics observed originated from profiles analyzed using the specialized protocol with hydraflock swabs.  As with the other secondary diagnostics, elevated variance parameters do not necessarily invalidate results.   Conventional methods resulted in significantly increased sensitivity compared to Rapid data, ultimately resulting in improved, and more inclusionary likelihood ratios (LRs) for known donors.  Increases of up to 16 orders of magnitude were observed for the mixture donors while single source blood samples resulted in differences of ~2-3 orders of magnitude once dilution factors of 1:20 and 1:30 were reached.

Authors

• Kaitlin Huffman (Federal Bureau of Investigation Laboratory, United States of America)
• Lori Flanagan (Federal Bureau of Investigation Laboratory, United States of America)
• Lilliana Moreno (Federal Bureau of Investigation Laboratory, United States of America)