The California Court of Appeals has affirmed a lower court decision in the murder conviction of Alvin Larry Davis, ruling that DNA evidence produced through the use of STRmix™ forensic software “is generally accepted as reliable by the relevant scientific community.” (C089567; Super.Ct.No.STK-CR-FE-2016-0004780)
STRmix™ – sophisticated forensic software capable of resolving mixed DNA profiles that previously were regarded as too complex or degraded to interpret – was used in the case to link Davis to a bloodstained shoelace found next to the body of the victim, 84-year-old Hazel Dingman.
Davis was sentenced to life in prison with no possibility of parole after being found guilty of first degree murder and and other offences by the lower court. On appeal, defense argued that evidence generated by STRmix™ and expert testimony that relied on STRmix™ to interpret DNA evidence should have been excluded under the test for the admission of new scientific evidence established by the California Supreme Court in People v. Kelly (1976 17Cal.3d24).
The Court of Appeals, however, rejected the defendant’s argument, noting that “establishing reliability is the overriding factor when a party seeks to admit evidence based on a new scientific technique.” Citing People v. Leahy (1994, 8 Cal.4th587, 612), the Court ruled that general acceptance means “a consensus drawn from a typical cross-section of the relevant, qualified scientific community. Unanimous acceptance is not required; rather, the test is met if use of the technique is supported by a clear majority of the members of that community.”
The Court of Appeals’ opinion went on to state that there was nothing in the record to suggest that the expert testimony provided during the lower court trial did not fairly and impartially assess the position of the relevant scientific community with regard to the general acceptance of the STRmix™ method.
The Court also rejected the defendant’s contention that the “black box” proprietary nature of STRmix™ prohibits the broader scientific community from fully assessing the software’s reliability. Affirming that a defendant in a criminal matter may obtain access to the STRmix™ source code under a non-disclosure agreement, the Court declared, “… a scientific technique or method need not be subject to testing and/or assessment by the scientific community at large to satisfy the requirement of general acceptance.”
Acknowledging that STRmix™ “has been used for DNA analysis since 2012 and is widely used by forensic laboratories across the world,” the decision pointed out that testimony presented at the hearing “established that the STRmix™ method has been subjected to extensive empirical testing and found to be accurate and reliable by the FBI and numerous forensic laboratories.”
STRmix™ is currently being used by 69 local, state, federal, and private forensic laboratories throughout the U.S., as well as all nine state and territory forensic laboratories in Australia and New Zealand, and 14 labs in other countries including the U.K., Ireland, Canada, Finland, Switzerland, and Denmark. In the decade since it was introduced, STRmix™ has been used to interpret DNA evidence in more than 300,000 cases worldwide.
STRmix™ works by assessing how closely multitudes of proposed DNA profiles resemble or can explain an observed DNA mixture. Relying on proven methodologies routinely used in computational biology, physics, engineering, and weather prediction, STRmix™ calculates the probability of observed DNA evidence by assuming the DNA originated from either a person of interest or an unknown donor. These two probabilities are then presented as a likelihood ratio (LR), inferring the value of the findings and level of support for one proposition over the other.
The success STRmix™ has enjoyed to date led the team behind its development to introduce two related products which when used in combination with STRmix™ complete the full workflow(external link) from analysis to interpretation and database matching:
- FaSTR™ DNA, expert forensic software which rapidly analyzes raw DNA data generated by genetic analyzers and standard profiling kits and assigns a number of contributors (NoC) estimate; and
DBLR™, an application which when used with STRmix™ allows forensic laboratories to calculate millions of LRs in seconds, visualize the value of their DNA mixture evidence, and carry out mixture-to-mixture matches, now allowing kinship analysis.