TrueAllele® Featured On Dateline Episode in Bucket Hat Case

Back to Presentations

Computer interpretation of RapidHIT mixtures for DNA match information

Paul Berry, MS, Baron Andrus, MS, William Allan, MS, Mark Perlin, PhD, MD, PhD, "Computer interpretation of RapidHIT mixtures for DNA match information", American Academy of Forensic Sciences Annual Scientific Conference, New Orleans, LA, 13-Feb-2026.


Talk

PowerPoint presentation of the conference talk.

PDFDownload Handout
PPTDownload PowerPoint

Abstract

Learning Overview: The RapidHIT™ instrument (Applied Biosystems™) facilitates rapid genotyping of evidentiary samples, providing timely investigative leads. However, its limitations present challenges in the interpretation of DNA mixtures. When used in conjunction with TrueAllele® probabilistic genotyping software (Cybergenetics), statistical weight can be assigned to potential matches, offering scientifically supported match probabilities even for complex or low-level DNA profiles generated by the RapidHIT ID system.

Impact Statement: This presentation will enhance the forensic science community’s understanding of how the integration of RapidHIT™ technology with TrueAllele® probabilistic genotyping software can improve the interpretation of complex Rapid DNA results. While RapidHIT provides fast genotyping from evidentiary samples, its limitations in mixture resolution or low-level single source profiles can hinder effective analysis. By incorporating TrueAllele’s statistical approach to mixture deconvolution and match probability calculation, forensic practitioners can make more informed, scientifically defensible decisions in time-sensitive situations.

RapidHIT™ ID (Applied Biosystems™) is a Rapid DNA technology. The instrument transforms input DNA into short tandem repeat (STR) data in about 1.5 hours. Rapid machines were originally intended for suspect identification from saliva reference samples in police stations. Technology advancements have expanded Rapid's scope to less pristine DNA items [1].

Most DNA evidence is a mixture of two or more people. STR mixture data are often difficult to interpret by conventional human or computer review, due to allele overlap and polymerase chain reaction (PCR) allele artifacts such as stutter, imbalance, and dropout. Mixtures can be highly probative, particularly when they place multiple people together at a crime scene.

While DNA laboratory protocols can mitigate PCR artifacts, Rapid technology often cannot. Rapid's heterozygote allele imbalance may exceed 40%, showing pronounced allele dropout [1]. Indeed, Rapid ID can signal PCR stochastic effects (STO flag). The Louisiana State Police (LSP) finds human review of Rapid data unsuitable for DNA mixtures, determining them to be uninterpretable.

Cybergenetics invented TrueAllele probabilistic genotyping (PG) 25 years ago [2]. The technology uses Bayesian probability to mathematically separate genotypes from DNA mixture data, one for each contributor. Comparing separated probabilistic genotypes produces a reliable likelihood ratio (LR) match statistic that measures DNA association strength [3]. The LSP uses TrueAllele for DNA mixture interpretation [4].

The LSP designed a DNA validation study to assess Rapid ID on single-source and two-person mixtures. The mixture ratios were roughly 1:9 (10%), 1:4 (20%), and (1:1) 50%. The laboratory used saliva samples from known individuals to synthesize the mixtures. Two swab types were compared, cotton and flocked.

LSP tested the DNA using RapidINTEL Plus Sample Cartridges (Applied Biosystems™) containing 21 autosomal STR loci. The "General" protocol was applied to cotton swabs, and "Specialized" (low-template) to flocked swabs. Three different Rapid machines were used, producing triplicate data for each mixture condition. Stochastic PCR variation was observed on both single-source and mixture STR data.

Cybergenetics ran TrueAllele on the Rapid mixture data. Using short triplicate computer runs (5,000 MCMC readout cycles each), a swab's total interpretation time was about 30 minutes. The major contributor was easily identified, with log(LR) match information generally at 25 ban. The 10% to 20% minor components averaged around 15 ban. Minor components of 1% to 5% yielded around 5 ban. The Specialized-flocked protocol was more informative with extremely low mixture amounts.

Our study showed that TrueAllele interpretation of Rapid ID two-person saliva mixtures was fast and informative. No instrument or cartridge calibration was needed, since TrueAllele models the PCR experiment directly from the sample's STR evidence data. These results demonstrate the potential for timely and highly informative Rapid analysis of DNA mixtures.

References

  1. Kowalczyk, E., King, J., Lagacé, R., Cihlar, J. "Utilization of the Rapid RapidINTEL™ Plus Cartridge on the RapidHIT ID Instrument", Promega's Thirty Fifth International Symposium on Human Identification, San Antonio, TX, 2024
  2. Perlin, M.W., Szabady, B. Linear mixture analysis: a mathematical approach to resolving mixed DNA samples. Journal of Forensic Sciences, 46(6):1372-7, 2001.
  3. Perlin, M.W., Legler, M.M., Spencer, C.E., Smith, J.L., Allan, W.P., Belrose, J.L., Duceman, B.W. Validating TrueAllele® DNA mixture interpretation. Journal of Forensic Sciences, 56(6):1430-47, 2011.
  4. Legler, M., Shawhan, Z., Naylor, J., Berry, P. "Solving major crime from minor contributors: A case report using TrueAllele® Casework", Promega's Thirty Fourth International Symposium on Human Identification, Denver, CO, 20-Sep-2023.

Links


  • American Academy of Forensic Sciences 78th Annual Scientific Conference - Program
Back to top