In April of 2010, the Scientific Working Group on DNA Analysis Methods (SWGDAM) published a report entitled Interpretation Guidelines for Autosomal STR Typing by DNA Testing Laboratories. This document updates and supersedes the previous SWGDAM Interpretation document released ten years earlier in 2000 and is now being used as guidance in crime labs across the nation.
For anyone in the legal or forensic DNA profession, it is extremely important to be aware of the changes, especially in DNA mixture interpretation, that have come about because of this document. These changes are having a direct impact on current DNA cases, and, although not intended to apply retroactively, the new guidelines may play a role in cases involving mixtures from years past that have not yet gone to trial.
DNA mixture interpretation has been a highly contentious subject for many years with some noted authorities in the DNA field commenting that if you give 10 analysts a mixture to interpret, you will likely end up with 10 different answers. With the increase in the number of "touch" or trace DNA samples, the mixtures that DNA analysts are being asked to interpret have grown even more complex. Instead of simply dealing with the typical two-person mixtures inherent with sexual assault cases, analysts are now seeing mixtures involving three, four, five, or more, persons obtained from items as diverse as knife handles, screwdrivers, steering wheel swabs, and clothing items. In addition, the DNA obtained from these touch items may be of low quantity and quality - both of which make mixture interpretations all the more difficult.
The purpose of the 2010 SWGDAM document is to provide guidance on interpreting mixtures and applying statistics to those mixtures. The guidelines were designed to help with consistency among laboratories and among individual analysts within a single laboratory.
What are the new guidelines and what effect are they having? One new guideline explicitly states that the mixture must be interpreted prior to comparison with known standards. This was (and is) not always common practice in some laboratories. The goal of this guideline is to help eliminate potential bias on the part of the analyst as it has been noted that it is possible for an analyst, who will often work closely with police and prosecutors, to have some form of "observer bias" - defined as "interpreting data in a manner consistent with one's expectations". Other new guidelines include the establishment by the lab of an Analytical Threshold, (defined as the minimum height requirement at and above which detected peaks can be reliably distinguished from background noise) and the Stochastic Threshold, "the peak height value above which it is reasonable to assume that, at a given locus, allelic dropout of a sister allele has not occurred". These thresholds are very important when determining which loci can be used in the statistical calculation and which cannot. For example, guideline 4.6.3 states that "loci with alleles below the stochastic threshold may not be used for statistical purposes to support an inclusion". And, standard 5.1 dictates that "a statistical calculation meant to represent all possible contributors to a mixture (such as CPI/CPE) should not be performed at a locus when there is a reasonable possibility that allele dropout could have led to the loss of an entire genotype". For some labs, these guidelines won't affect their conclusions as they have always been doing things this way. For many other labs, however, following the new guidelines will have a major impact on their statistical conclusions. For example, one recent case I reviewed reported a Combined Probability of Inclusion (CPI) statistic of 1 in 1.5 million for the Caucasian population. During the time period between when the case was completed, and when the case went to trial, the laboratory's mixture interpretation guidelines had been updated to come into line with the SWGDAM guidelines. When the statistic was recalculated (by the same laboratory, at the request of the prosecution) the conclusion for the Caucasian population was vastly different - the probability was now determined to be 1 in 1,400!
The 2010 SWGDAM guidelines also state that labs must perform statistics in "support of any inclusion that is determined to be relevant in the context of the case". This may seem like a no-brainer, but some labs have not routinely reported statistics on samples with very low-level data present, or with only a few alleles present. They would, however, make a statement to the effect that the suspect, "could not be excluded from" the evidence item. The new guidelines ensure that if a lab is including a person as a contributor (which is the same thing as stating that one "cannot be excluded"), they must give weight to that inclusion by reporting a statistic - even if the stat is nearly meaningless.
The SWGDAM guidelines are not considered "standards" and laboratories will not be audited against the document, nor are they even required to follow it. However, most laboratories will likely change, or have changed, their mixture interpretation procedures to fall in line with the SWGDAM document. In addition, the document is not intended to apply retroactively. However, some attorneys who have examined this issue have come to the conclusion that this is not enforceable. For example, Jules Epstein, Associate Professor of Law at Widener University School of Law gave a presentation to attendees at the NFSTC Mixture Interpretation Workshop held in March 2011 entitled, "The Law, Ethics, and DNA Interpretation". He brought up several interesting points in his presentation including the fact that considering that these are "new guidelines", does that mean that the old way of doing things was not good enough? Upon his questioning, at least one attendee at the meeting (the vast majority of whom were practicing forensic scientists) commented that, "This is the way the community should have been handling mixtures all along".
The bottom line is that labs may have been over-estimating the rarity of certain DNA profiles when following the old SWGDAM guidelines which offered very little guidance on the topic of mixture interpretations to begin with. Epstein stated that there may be possible Brady implications to this issue. Some laboratories are trying to address the Brady implications by notifying all the prosecutors they serve of the updated mixture interpretation guidelines and even offering to recalculate mixture statistics on older cases using the new guidelines. The possibility of admissibility hearings regarding the new guidelines and the new methods to calculate and report mixture statistics was also raised.
With the advent of the new guidelines, it is likely that more "Inconclusive" results will be reported, where in the past, the conclusion might have been "cannot be excluded". In addition, the statistical results for many of the mixtures obtained by DNA labs will likely be less rare, i.e., it will be more likely to find another individual who could also have been a contributor. Overall, the guidelines should make the statistical conclusions more conservative. However, it is highly likely that cases reported in the past will not undergo any re-calculation of statistical results as the attorneys involved may not even be aware of the changes addressed under the new SWGDAM guidelines or their implications.
Suzanna Ryan, MS, D-ABC, is a former forensic DNA analyst and forensic DNA Technical leader with 15 years of experience in the field of Forensic Serology and DNA Analysis. She has had the opportunity to work for both public and private DNA laboratories and has testified numerous time for both the prosecution and the defense.
©Copyright - All Rights Reserved
DO NOT REPRODUCE WITHOUT WRITTEN PERMISSION BY AUTHOR.
