The Daily Bulletin Archive

Please join the Chemistry and Physics departments for the seminar "STRmix: New Software for Complicated DNA Mixtures," presented by Matthew Pothier, candidate for the master of science degree in forensic science at Buffalo State, on Thursday, November 9, from 12:15 to 1:30 p.m. in Science and Mathematics Complex 169. All students, faculty, and staff are welcome.

STRmix is a new software that is being deployed globally to laboratories that conduct analysis on DNA samples and mixtures. Developed in New Zealand by the Institute of Environmental Science and Research (ESR) and Forensic Science South Australia (FSSA), STRmix uses a continuous probability genotyping model to interpret DNA samples and mixtures. Currently there are several different types of DNA software and “in-house” counterparts used in crime laboratories across the globe; however, STRmix is advertised as the best for solving unresolvable mixed DNA profiles when low concentrations and high number of contributors are present. This presentation will go into detail of the basics of DNA analysis, covering how STRmix software works along with validation studies and several cases in which it was involved.

The first study covered in my seminar will be the ESR validation of STRmix for forensic casework. The purpose of this study is to determine if the software meets the SWGDAM Guidelines for the Validation of Probabilistic Genotyping Systems. The paper establishes a series of controlled DNA samples that undergo the analysis of STRmix with respect to several conditions such as sensitivity and specificity. The ESR determined that the software is capable of performing the interpretation of a single source and mixed DNA samples. The second study will be that of the validation done by the FBI. The results of the FBI’s internal validation concluded that STRmix provides a number of advantages over the previous DNA analysis software and can be used in forensic laboratories. The third study is that of a court case in Niagara County in which STRmix was used for the first time. This paper was chosen to show what legal obstacles a crime laboratory has to overcome when introducing a new software for analyzing DNA and how STRmix meet those required standards. The fourth and final study is an actual court case in Texas in which STRmix was used to analyze a DNA sample that previous fell in the drop-out range and was not identified. With the application of a continuous probability genotyping model, evidence that could not include the suspect was later able to identify the suspect with a likelihood ratio of 151 octillion. The weight of the DNA evidence along with the forensic artist sketch and testimony of witnesses led to a conviction.