Pennsylvania prosecutors use TrueAllele in homicide guilty plea

Back to Presentations

Compute first, ask questions later: an efficient TrueAllele® workflow

M. Bowkley and M.W. Perlin. "Compute first, ask questions later: an efficient TrueAllele® workflow", Midwestern Association of Forensic Scientists 43rd Annual Fall Meeting, St. Paul, MN, 9-Oct-2014.


Talk

PowerPoint presentation with live audio recording of Mr. Bowkley's talk.



Download Handout
PPTDownload PowerPoint


Abstract

Today's crime laboratory can rapidly progress DNA from item to electropherogram (EPG) in just a few days. And then the data review begins. Inspecting data peaks, applying rules, and following procedures can take considerable time on DNA mixtures or low-level samples. Tedious data review is then followed by more demanding tasks, such as genotype inference or likelihood ratio (LR) calculation.

But suppose the mixture analysis, genotype inference and match statistics were all done before ever reviewing the data. Then the DNA analyst would have a wealth of information at their disposal when examining the evidence. Data review would become a small part of the overall forensic analysis process.

Cybergenetics TrueAllele® Casework provides this workflow capability.


  • An analyst starts with a full 96-well plate of EPG data files (.fsa or .hid).
  • Their TrueAllele workstation automatically analyzes EPGs for peak size and height information, and uploads this data to a TrueAllele database server.
  • Overnight, the computer solves all the mixture problems for up to three (or more) contributors, inferring genotypes and writing them onto the database.

The next morning, the analyst has the STR data, mixture weights and separated probabilistic genotypes all available for immediate visual review on their TrueAllele workstation. Comparison of genotypes (evidence or reference) provides instantaneous LR calculation, with a user settable theta. A positive log(LR) value indicates inclusion, while a negative value indicates exclusion.

With challenging data, more analysis may be needed on some samples (e.g., longer computer runs, using known references). By computing first, the analyst can ask TrueAllele interesting forensic questions later on, always in the context of visually presented computer answers. This automated process provides a fast and easy workflow for an analyst to fully solve their DNA mixture problems.

This "compute first" workflow integrates with TrueAllele's investigative database DNA matching capability. Because the TrueAllele DNA Database has high specificity (unlike CODIS), all mixture genotypes are uploaded. The database automatically compares the newly inferred genotypes with all the lab's (new and) old genotypes, determining LR values. Evidence can be compared with evidence (case-to-case matches) and reference (within-case matches) items.

This talk describes TrueAllele's "compute first, ask questions later" workflow, and its use in a DNA crime laboratory. A case example will illustrate how it helps the DNA analyst.