Opinion
CRIMINAL ACTION 20-00091-BAJ-RLB
11-01-2023
UNITED STATES OF AMERICA v. DAMOND REYNARD LOCKETT
RULING AND ORDER
BRIAN A. JACKSON UNITED STATES DISTRICT COURT JUDGE.
Defendant Damond Reynard Lockett is charged with one count of possessing with intent to distribute 100 grams of heroin and fentanyl, and three related gun charges. Each gun charge results from Defendant's alleged possession of a single (stolen) firearm. To establish Defendant's possession of the firearm, the Government seeks to introduce DNA identification evidence generated by TrueAllele, a computer software program that analyzes complex DNA mixtures using statistical modeling. Now, Defendant challenges the admissibility of the proposed TrueAllele evidence, arguing that it is unreliable because it has an unacceptable error rate, has not been subjected to adequate peer review, and is not generally accepted in the relevant scientific community. Additionally, Defendant asserts that “the science[] ... is hard to understand and confusing.” The Government opposes Defendant's motion, asserting that TrueAllele and the methodology it employs are sufficiently reliable to pass muster under Rule 702 of the Federal Rules of Evidence and Daubert v. Merrell Dow Pharmaceuticals, Inc., and that any “confusion” will be mitigated by deliberate questioning of its expert witness. For reasons below, Defendant's Motion will be denied.
I. RELEVANT FACTS AND PROCEDURAL HISTORY
This case presents a tortured procedural history with numerous substantive motion filings, evidentiary issues, hearings, and continuances of the trial date. To follow are the details most relevant to this Motion.
On November 19, 2020, the grand jury returned a three-count indictment charging Defendant with possession with the intent to distribute heroin and fentanyl, possession of a firearm in furtherance of a drug trafficking crime, and possession of a firearm after previously being convicted of a felony offense. (Doc. 1). The two firearms offenses related to Defendant's alleged possession of the same gun: a Smith & Wesson Sigma SD40VE .40 caliber semiautomatic pistol. (Id.). The drugs and the gun were recovered in May 2020, when officers of the East Baton Rouge Parish Sherriff's Office Narcotics Division executed two search warrants of residences associated with Defendant, following 10 days of electronic surveillance of Defendant's activities. (See Doc. 30 at 1-3).
Trial was originally set to begin on January 13, 2021. (Doc. 10). On December 17, 2020, however, the Court granted Defendant's motion to declare this case complex, due to the volume and variety of discovery produced by the Government- which (at the time) included surveillance video, police reports, video interviews, search warrant applications, crime lab reports, and a Title III wiretap in an unrelated matter-as well as the potential for additional charges based on certain unindicted activities. (Docs. 13, 14). In the same December 17 Order, the Court continued the original trial date and related deadlines. (Doc. 14 at 2).
In the months that followed, Defendant filed an unsuccessful motion to suppress the drug and gun evidence against him, (see Docs. 19, 30), and four successful motions to continue his trial, (see Docs. 32, 38, 43, 45). The last of Defendant's motions to continue resulted in trial being re-set to December 14, 2021. (Doc. 47).
On November 19, 2021, the Government took its turn seeking a continuance. (Doc. 49). In its November 19 motion, the Government stated that on November 11, 2021, while preparing for trial, it learned for the first time “that a DNA swab ... taken on the firearm forming the basis of [the gun] charges, as well as a reference swab from defendant, was [sic] never submitted to the Louisiana State Police Crime Lab for testing,” and that due to a “substantial backlog” testing could not be completed until “early 2022.” (Doc. 49 ¶¶ 2-3). The Government asserted that this DNA evidence was “instrumental to an exoneration or a conviction in this matter,” and that “[j]ustice cannot be served without the inclusion of this evidence at trial.” (See Doc. 49 ¶ 4). Defendant objected to yet another continuance, (id. ¶ 8), and the Court denied the Government's motion to continue concluding that the Government failed to exercise due diligence to obtain the Crime Lab's analysis of its DNA evidence. (See Doc. 51 at 2-3) (citing United States v. Burrell, 634 F.3d 284, 290 (5th Cir. 2011)).
Still, trial did not go forward on December 14, 2021, this time because the Government learned during its trial preparation that a potential witness in its casein-chief was previously represented by the Federal Public Defender's Office, Defendant's appointed counsel at the time. (Doc. 58). Following a hearing, the Court agreed that a potential conflict existed, ordered that Defendant be appointed new counsel, and re-set trial to June 27, 2022.
Meantime, benefitting from the additional delay, on January 26, 2022, the Government obtained the LSP Crime Lab's initial analysis of the DNA swabs. (Doc. 95-6 at 2-3). The Crime Lab's January 26 report was inconclusive “[d]ue to the limited nature of the [DNA] profile” obtained from the handgun. (Id. at 2). The Government produced the January 26 Report to Defendant on March 10, 2022. (Id. at 1).
Not satisfied with the Crime Lab's results, the Government requested “a more enhanced test” of the DNA sample obtained from the firearm, this time using TrueAllele statistical modeling. (See Doc. 96-3 at 2). On June 14, 2022-two weeks before trial-the Crime Lab issued its TrueAllele report, which concluded that Defendant “cannot be excluded as a contributor to [the DNA] profile” recovered from the handgun, and that a match between the firearm sample and the sample provided by the Defendant was 109 billion times more likely than a coincidental match between the firearm sample and “an unrelated ... African American.” (Doc. 96-2 at 2). According to the Crime Lab's June 14 TrueAllele report, this “likelihood ratio” indicates “Very Strong Support” for the conclusion that Defendant's DNA is included in the profile recovered from the handgun. (Id.).
The Government immediately forwarded the TrueAllele report to Defendant's counsel. (Doc. 96-3). Shortly thereafter, on June 20, 2022, Defendant moved to exclude the TrueAllele report from trial, arguing that its late production violated Federal Rule of Criminal Procedure 16 and this Court's Scheduling Order. (Doc. 95). The Court denied Defendant's June 20 motion, determining that any prejudice resulting from the Government's late production of the TrueAllele report could be cured by yet another continuance, to allow Defendant the opportunity to retain an expert to review and challenge the TrueAllele analysis. (Doc. 100). Thereafter, upon conferring with the parties, the Court re-set trial to October 31, 2022. (Doc. 106).
Still more delays followed. On September 28, 2022, the Government obtained a Superseding Indictment, charging Defendant with a fourth count-possession of a stolen firearm. (Doc. 109). Notably, the alleged stolen firearm is the same Smith & Wesson that forms the basis of the two original gun charges. (Id.). There is no explanation for why the Government waited two years to add this wrinkle to the case.
After the Superseding Indictment was filed, the Court re-set trial to February 13, 2023. (Doc. 119). On January 25, 2023, Defendant filed the instant Daubert motion challenging the admissibility of any testimony or evidence related to the TrueAllele report. (Doc. 134). On February 10, 2023, the parties submitted a joint notice clarifying that Defendant's Daubert challenge is limited solely to the following aspects of the TrueAllele analysis:
The bases for this challenge are as follows: The defendant is seeking to challenge the reliability of the TrueAllele science, specifically as to high allele sharing, its peer review and its general acceptance in the relevant scientific community. Additionally, the defendant is challenging the science's admissibility as he believes it is hard to understand and confusing. This supersedes the arguments made in the motion previously filed (Rec. Doc. 134).(Doc. 141 ¶ 2). The parties further stipulated that “defendant does not object to the TrueAllele application in this case by the Louisiana State Police Crime Laboratory or to Dr. Naragoni's qualifications as an expert [for the Government] in Forensic DNA Analysis.” (Id. ¶ 3).
Based on Defendant's Daubert challenge, the Court set a Daubert hearing for April 3, 2023, and continued trial without date. (Docs. 140, 147). The Daubert hearing featured a single witness, called by the Government: Paul Berry, the DNA Technical Leader of the LSP Crime Lab. (Doc. 148). In other words, Defendant did not retain an expert or produce any witnesses to support his Daubert challenge. At the hearing's conclusion, the Court ordered additional briefing from the parties and took the matter under advisement. (Id.; see Docs. 151, 152). This Order follows.
II. LAW AND ANALYSIS
A. Standard
The admissibility of expert testimony is governed by Federal Rule of Evidence (“Rule”) 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), which require the Court to serve as a gatekeeper and ensure that all scientific testimony is relevant and reliable. This gatekeeping role extends to all expert testimony, whether scientific or not. See Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 147 (1999).
“Under Rule 702, the court must consider three primary requirements in determining the admissibility of expert testimony: 1) qualifications of the expert witness; 2) relevance of the testimony; and 3) reliability of the principles and methodology upon which the testimony is based.” Fayard v. Tire Kingdom, Inc., No. 09-cv-171, 2010 WL 3999011, at *1 (M.D. La. Oct. 12, 2010) (Jackson, J.). Here, again, Defendant does not challenge the qualifications of the Government's proposed expert, or the relevance of TrueAllele testimony to his case. (Doc. 141 ¶ 3).
The Court's consideration of Defendant's challenge to the reliability of TrueAllele statistical modeling is guided by the factors set forth in Daubert:
These factors include: (1) whether the technique in question has been tested; (2) whether the technique has been subject to peer review and publication; (3) the error rate of the technique; (4) the existence and maintenance of standards controlling the technique's operation; and (5) whether the technique has been generally accepted in the scientific community.United States v. Perry, 35 F.4th 293, 329 (5th Cir. 2022) (quotation marks, alterations, and citations omitted), cert. denied 143 S.Ct. 462 (2022); accord Watkins v. Telsmith, Inc., 121 F.3d 984, 989 (5th Cir. 1997). This list is merely illustrative, however, and the Supreme Court has emphasized that “the Daubert analysis is a ‘flexible' one, and that ‘the factors identified in Daubert may or may not be pertinent in assessing reliability, depending on the nature of the issue, the expert's particular expertise, and the subject of his testimony.” Pipitone v. Biomatrix, Inc., 288 F.3d 239, 244 (5th Cir. 2002) (citing Kumho Tire, 526 U.S. at 150). “The district court's responsibility is ‘to make certain that an expert, whether basing testimony upon professional studies or personal experience, employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field.'” Id. (quoting Kumho Tire, 526 U.S. at 152).
Ultimately, the Court has broad discretion when deciding whether to admit expert testimony. See Perry, 35 F.4th at 329. At same time, the Court remains cognizant that “the rejection of expert testimony is the exception and not the rule,” even when the testimony at issue may be “potentially misleading or confusing.” See Perry, 35 F.4th at 329 (citing authorities); accord Barnett v. Nat'l Cont'l Ins. Co., No. 17-cv-153, 2019 WL 126732, at *3 (M.D. La. Jan. 8, 2019) (deGravelles, J.). “Vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence.” Perry, 35 F.4th at 330 (quoting Daubert, 509 U.S. at 596).
B. Discussion
As stated, Defendant limits his Daubert challenge to three factors: whether TrueAllele statistical modeling is adequately peer reviewed; whether its error rate is unacceptable in instances of “high allele sharing”; and whether it is generally accepted in the relevant scientific community. (Doc. 141 ¶ 2). Additionally, Defendant contends that TrueAllele evidence should be excluded because “it is hard to understand and confusing.” (Id.).
i. Background on TrueAllele
To put Defendant's arguments in context, the Court provides a brief overview and description of TrueAllele analysis. The following information is drawn from the declaration and published articles of Dr. Mark W. Perlin, TrueAllele's creator and the founder of Cybergenetics, Inc. (the private company that owns and operates the program), (Doc. 139-1, “Perlin Decl.”), and recent caselaw addressing the admissibility of TrueAllele analysis at trial.
“TrueAllele is a probabilistic genotyping computer system that interprets DNA evidence using a statistical model.” (Perlin Decl. ¶ 4). Genotype is the genetic composition of a cell or an individual, and may serve as a unique “genetic bar code” in forensic applications. (See Doc. 151-1 at p. 2). When a single person's DNA produces unambiguous genetic data, a genotype can be “readily determined.” (Perlin Decl. ¶ 6). But when the genetic data is less definitive, or when two or more people contribute genetic material to the same DNA sample, “uncertainty arises.” (Id. ¶ 7). Such “genotype uncertainty may translate into reduced identification information”-i.e., lead to inconclusive results-when (as here) traditional DNA analysis is used to compare a genetic sample swabbed from a firearm, and a genetic sample swabbed from the defendant. (Id., ¶ 9).
This is where TrueAllele steps in.
According to Dr. Perlin, in cases where human review might be less reliable or not possible-for example, when a DNA sample contains DNA from multiple individuals or when its properties have been affected by external elements prior to collection-TrueAllele can be used to determine the probability that the sample matches the DNA of a given person. For this, TrueAllele employs a two-step identification process: the software first infers genotypes contained in the sample using evidence data, and then second, matches genotypes by comparing evidence with a suspect relative to a population.United States v. Anderson, __F.Supp.3d__, 2023 WL 3510823, at *1 (M.D. Pa. May 17, 2023) (Brann, C.J.) (citing Perlin Decl.; quotation marks, alterations, and footnotes omitted). The strength of association between the inferred genotypes and the suspect's genotype is then reported as a single number-the “likelihood ratio”- where the numerator is the probability (or likelihood) that the inferred genotypes from the forensic genetic sample (here, the firearm) are a match to the suspect's genotype, and the denominator is the probability (or likelihood) that the inferred genotypes from forensic genetic sample are a match to the genotype of a random person from the general population. (See id., ¶¶ 10.2, 11).
See Mark W. Perlin, Joseph B. Kadane, & Robin W. Cotton, Match Likelihood Ratio for Uncertain Genotypes, 8 LAW, PROBABILITY, & RISK 289, 292 (2009).
In the instant case, TrueAllele statistical modeling concluded that a match between the firearm sample and Defendant's sample was 109 billion times more likely than a coincidental match between the firearm sample and a sample drawn from another African American. (See Doc. 96-2 at 2).
ii. Analysis
Probabilistic genotyping-of the type performed by TrueAllele-is not new. It is, however, a relatively recent addition to the Government's toolkit, and there is no controlling authority from the U.S. Court of Appeals for the Fifth Circuit addressing its admissibility as evidence at trial. Farther afield, however, recent opinions from the Sixth Circuit and the U.S. District Court for the Middle District of Pennsylvania hold that probabilistic genotyping-whether performed by TrueAllele or a competitor program-satisfies Rule 702 and the Daubert reliability factors. See United States v. Gissantaner, 990 F.3d 457, 463-67 (6th Cir. 2021) (district court erred by excluding probabilistic genotyping evidence performed by STRmix-one of TrueAllele's competitors-where the evidence established that “it is the product of reliable principles and methods” (discussing authorities)); Anderson, 2023 WL 3510823, at *1 (rejecting defendant's Daubert challenge to TrueAllele: “TrueAllele has been tested and validated, subjected to peer review, and broadly accepted in the field of forensic science.” (discussing authorities)); but see Morten v. State, 215 A.3d 846, 875 (Md.App. 2019) (reversing murder conviction based, in part, on TrueAllele analysis linking the defendant to the murder weapon because even if TrueAllele evidence is admissible “as a matter of law,” at trial the defendant was “erroneously prohibited from challenging the TrueAllele test results that linked him to the ostensible murder weapon”). The Court's analysis of Defendant's challenge is guided by these well-reasoned authorities.
The Government represents in its post-hearing brief that “TrueAllele has been admitted in courts over 120 times since 2009.” (Doc. 151 at 1; see also Perlin Decl. ¶ 19). Defendant does not contest this statement, but directs the Court's attention to State v. Briscoe, a September 2022 decision from the Twenty-Fourth Judicial District Court for the Parish of Jefferson, Louisiana, where the trial court excluded TrueAllele evidence from the trial of a six-year-old murder case. (See Doc. 134-2). Certainly, in Briscoe, the trial court expressed concern regarding “the reliability of this specific program [TrueAllele].” (Id. at 5). Ultimately, however, the trial court's ruling was grounded on (1) the State's last-minute disclosure of the TrueAllele evidence, (id. at 6 (“This Court does not think it is ‘fair' to these Defendants to bring this evidence AT THE LAST MINUTE .” (emphasis in original)); and (2) the State's inability to put together a compelling presentation at the Daubert hearing, (id. at 7 (“If the State was not prepared to present this evidence, it should not have attempted to introduce it.”)). The same concerns are not at issue here (despite the State's late disclosure) where each side has been afforded the opportunity to capably vet the proposed TrueAllele evidence.
Defendant's challenge to the second Daubert factor-whether TrueAllele is adequately peer reviewed-is quickly dispatched. In fact, Defendant concedes that TrueAllele is the subject of multiple peer-reviewed validation studies-eight to be exact, including five published in the Journal of Forensic Sciences. (See Doc. 151 at 5-10). Nonetheless, Defendant objects that the value of these studies is diminished because Dr. Perlin-TrueAllele's creator-“participated either directly or indirectly in the majority of [them].” (Doc. 152 at 2).
In Anderson, the Middle District of Pennsylvania rejected precisely the same argument, based on the same evidence submitted by the Government here, explaining that merely because “Dr. Perlin was the principal or supporting author of all but one of the peer-review publications” does not affect the outcome of the analysis because “‘this factor does not demand independent authorship,' as ‘peer review contains its own independence.'” See Anderson, 2023 WL 3510823, at ** 6-8 (quoting Gissantaner, 990 F.3d at 465 (citing authorities)). In Gissantaner, the Sixth Circuit reached the same conclusion as to STRmix-one of TrueAllele's statistical modeling competitors-explaining:
[The peer review] factor does not demand independent authorship- studies done by individuals unaffiliated with the developers of the technology. Independent studies, to be sure, advance the cause of reliability. But they are not indispensable. Peer review contains its own independence, as it involves anonymously reviewing a given experimenter's methods, data, and conclusions on paper. If experts have other scientists review their work and if the other scientists have the chance to identify any methodological flaws, that usually suffices. When scientific research is accepted by a reputable journal following the usual rigors of peer review, that represents a significant indication that it is taken seriously by other scientists, i.e., that it meets at least the minimal criteria of good science.Gissantaner, 990 F.3d at 465 (quotation marks and citations omitted)). The same reasoning applies with equal weight here. “Because TrueAllele has been the subject of multiple published peer-review studies, the factor weighs in favor of admissibility.” Anderson, 2023 WL 3510823, at *8.
Next, Defendant challenges the third Daubert factor-TrueAllele's error rate- specifically as it relates to instances of “high allele sharing.” (Doc. 141 ¶ 2). At the Daubert hearing, Defendant's counsel explained that “high allele sharing” occurs when the forensic swab collected from an item of evidence contains genetic material of “close relatives or other individuals that you may be related to,” and expressed generalized concerns “that the TrueAllele program does not accurately take into account for the [sic] high allele sharing.” (Doc. 150 at 10:2-12). Defendant's problem (for present purposes) is that the Government's unrebutted evidence at the Daubert hearing establishes the opposite-specifically, “that TrueAllele does not have a problem resolving mixtures where there are individuals who share alleles.” (Id. at 71:23-25). Citing a March 2014 Cybergenetics study-titled TrueAllele Casework Separates DNA Mixtures That Share Alleles-Mr. Berry testified that TrueAllele “does not have a problem parsing out” results from swabs containing a mixture of genetic material “regardless [of] how much allele sharing happens.” (Id. 72:10-13; accord 86:14-87:8). The study itself, introduced into evidence without objection at the Daubert hearing as Government's Exhibit 8, (Doc. 150 at 71:16-20), states the following conclusion:
Mixtures of related individuals can share more alleles at a locus than mixtures from unrelated individuals. Increased allele sharing between contributor genotypes can affect STR mixture interpretation, particularly human analysis methods based on STR peak height thresholds that discard allele quantity data. Since genotype modeling methods such as TrueAllele make better use of the quantitative data, and mathematically model the allele sharing of contributor genotypes, they can better separate the genotypes present in mixtures of relatives, and preserve DNA match information.
This study examined the computer's performance across low, middle, and high levels of allele sharing for both high and low template DNA amounts. Mixture weight was found to be reproducible, regardless of allele sharing level and DNA amount. DNA match information, measured as log(LR) match statistics, was reproducible for each allele sharing level and DNA amount. TrueAllele was sensitive and specific in all these mixture groups. With the exception of one group (high template, low sharing), there was no significant difference in TrueAllele's average match information across the low, middle and high allele sharing mixture groups for high or low template amounts.
TrueAllele is a useful computational tool that performs accurately and reliably on DNA mixtures where there may be higher allele sharing, such as mixtures of relatives. This study provides empirical support for TrueAllele's ability to resolve DNA mixtures that contain related individuals.
Absent any evidence to the contrary, Mr. Berry's testimony, coupled with the March 2014 Cybergenetics study, supports a finding that TrueAllele's error rate is calculable and consistent, even as it relates to instances of “high allele sharing.” This factor also weighs in favor of admissibility. Accord Anderson, 2023 WL 3510823, at *9 (“Dr. Perlin and the Government have presented evidence of validation studies showing how often TrueAllele falsely suggests a known noncontributor matches a DNA sample. In other words, experts at Cybergenetics and in independent crime labs have tested the reliability of TrueAllele's results by calculating an error rate. This factor therefore weighs in favor of admissibility.”).
Defendant's challenge to the fifth Daubert factor-TrueAllele's general acceptance in the relevant scientific community-is equally unpersuasive. At the Daubert hearing, Mr. Berry offered uncontradicted testimony that TrueAllele has been used in 46 states to date (including Louisiana), (Doc. 150 at 58:13-17), that half of all crime labs in the United States are currently using probabilistic genotyping technology, (id. at 55:8-18), and that “90 percent of the remaining laboratories [are] moving towards it,” and “will be using it” within the next five years, (id. at 55:8-18, 100:3-5). In Anderson, the Middle District of Pennsylvania considered similar evidence, ultimately concluding that “TrueAllele and the methods it employs are ‘generally accepted' by the relevant scientific community.” Anderson, 2023 WL 3510823, at *13. The Court reaches the same conclusion here. Again, this factor weighs in favor of admissibility. Id.
Finally, Defendant challenges TrueAllele's “admissibility as he believes it is hard to understand and confusing.” (See Doc. 141 ¶ 2). In support, Defendant cites a February 2023 case study (not peer reviewed) authored by Dr. William C. Thompson, Professor Emeritus of Criminology, Law, and Society at the University of California- Irvine, wherein Dr. Thompson “expressed some concerns about the manner in which the True Allele system use of [sic] the term ‘coincidental match.'” (Doc. 152 at 1).
The Fifth Circuit instructs that merely because expert testimony is “confusing,” or even “potentially misleading,” does not render it excludable under Daubert, provided that the testimony is otherwise admissible and subject to “the traditional tools of attacking the evidence”-i.e., effective cross-examination and contrary evidence aimed to resolve ambiguities and expose weaknesses. See Perry, 35 F.4th at 330 (citing authorities). Certainly, Dr. Thompson's opinions set forth in his case study are ample fodder for cross-examination. On this record, however, they are not a basis for exclusion.
In sum, for the reasons cited herein, and the additional reasons set forth by the Middle District of Pennsylvania in the Anderson decision, the Court finds that TrueAllele and the methodology it employs are sufficiently reliable, and admissible under Rule 702 and Daubert. In short, “TrueAllele has been tested and validated, subjected to peer review, and broadly accepted in the field of forensic science.” See Anderson, 2023 WL 3510823, at *1. As such, Defendant's motion must be denied.
III. CONCLUSION
Accordingly, IT IS ORDERED that Defendant's Motion (Doc. 134) challenging the admissibility of testimony regarding True Allele at trial be and is hereby DENIED.
IT IS FURTHER ORDERED that a telephone status conference be and is hereby SET for November 9, 2023 at 11:00 a.m. for purposes of selecting a new trial date and related deadlines. Dial-in instructions will be emailed to counsel prior to the conference.