Opinion
2573/14
01-12-2016
The People of the State of New York v. Donsha Carter, Defendant.
For the People: Maryam N'Ha Margo Hadden, Esq. For Defendant: Pauloma Martinez, Esq.
For the People: Maryam N'Ha Margo Hadden, Esq. For Defendant: Pauloma Martinez, Esq. Robert A. Schwartz, J.
Defendant stands indicted on charges of attempted murder in the second degree, assault in the first degree, criminal possession of a weapon in the second degree, criminal possession of a weapon in the third degree, and criminal possession of a weapon in the fourth degree. In support of their case, the People intend to introduce DNA evidence linking defendant to these crimes. Defendant has filed a motion seeking to preclude the People from calling at trial an expert witness to testify about conclusions reached through the use of the Forensic Statistical Tool (FST) — the computer software program used by the Office of the Chief Medical Examiner (OCME) to determine the likelihood ratio that defendant's DNA was a contributor to the DNA mixture recovered from a sweatshirt allegedly worn by the perpetrator. Defendant claims that the methods and software used by the OCME to determine the likelihood ratio are not generally accepted as reliable by the relevant scientific community. Alternatively, defendant seeks a Frye hearing (Frye v. United States, 293 F. 1013 [D.C. Cir. 1923]), to determine whether the use of the FST is generally accepted in the relevant scientific community. Defendant also seeks preclusion of the evidence on the ground that it is not relevant.
In the event this court denies preclusion, defendant seeks an order requiring the prosecution to disclose the FST source code and software to the defense and order the prosecution, under section 240.20(1)(c) of the Criminal Procedure Law, to provide the defense with the electronic computer files that contain electronic data of the DNA testing performed in this case. In the alternative, defendant seeks leave to submit a subpoena duces tecum to the OCME requiring it to produce the files.
For the reasons stated below, defendant's motion is denied in its entirety.
Background
The People allege that on May 2, 2014, defendant shot Darnell Rowe four times before fleeing. According to the People, nearby surveillance video captured the incident and showed the assailant wearing a sweatshirt and a baseball cap. It captured the shooter running into an apartment still wearing the sweatshirt and hat, and then leaving the apartment minutes later wearing different clothing. Later that day, the police executed a search warrant for the apartment and recovered the sweatshirt and hat seen in the video, as well as a gun. The police submitted those items to the OCME to test for the presence of biological material. Defendant was apprehended in Pennsylvania in July 2014.
In a report dated August 11, 2014, the OCME indicated that both the sweatshirt and the hat contained DNA sufficient for testing. Specifically, a mixture of DNA from at least two people, including at least one major male contributor, was found in scrapings from the baseball cap. A mixture of DNA was also found in scrapings from the sweatshirt, but profiles of the contributors to the mixture could not be determined. The results were deemed to be suitable for comparison to a known sample, however. In a report dated February 5, 2015, the OCME indicated that the DNA found on the gun was insufficient for testing. In a decision and order dated April 21, 2015, the Honorable Leslie Leach ordered defendant to provide a DNA sample to the People for comparison to the DNA recovered from the sweatshirt and hat, and defendant did so.
The OCME created a DNA profile from defendant's sample and compared it to scrapings from the hat and from the hooded sweatshirt. The OCME found that defendant's DNA profile matched the DNA profile of the major contributor to the DNA found on the baseball cap, and that that DNA profile would be expected to be found in approximately one in greater than 6.80 trillion people. Using FST — a software program developed by the OCME to state likelihood ratios between the probability that a defendant is a contributor to a DNA mixture and that he is not — the OCME compared defendant's DNA profile to the mixture of DNA found on the sweatshirt and determined that defendant was a possible contributor to that mixture. Specifically, the OCME concluded that "[t]he DNA mixture found on the scrapings from the hooded sweatshirt is approximately 5640 times more probable if the sample originated from [defendant] and two unknown, unrelated persons than if it originated from three unknown, unrelated persons. Therefore, there is very strong support that [defendant] and two unknown, unrelated persons contributed to this mixture, rather than three unknown, unrelated persons." (July 6, 2015, OCME Laboratory Report).
Defendant is challenging the methods used by the OCME to determine the likelihood ratio that defendant was a contributor to the DNA mixture found in the sweatshirt versus that the contributors were three other unknown individuals. Defendant is not challenging the methodology by which OCME determined that defendant was the main contributor to the DNA found on the baseball cap. FST is not New, Novel or Experimental, and in any Event Meets the Frye Standard.
FST is an in-house computer program developed and validated by OCME that calculates likelihood ratios. It is approved by the New York State Commission on Forensic Science and its DNA subcommittee. A likelihood ratio is a statistical calculation that quantitatively expresses the likelihood that one scenario — the probability of seeing a mixture of DNA if the defendant did contribute to the mixture — exists over another — the probability of seeing the same mixture if the defendant did not contribute to the mixture, and only unknown persons did. OCME interprets likelihood ratios as follows: a likelihood ratio of 1.00 is inconclusive; a likelihood ratio in the range of 1.1 to 10 provides limited support for the proposition that the defendant was a contributor; a likelihood ratio in the range of 10 to 100 provides moderate support; a likelihood ratio in the range of 100 to 1000 provides strong support; and a likelihood ratio greater than 1000 provides very strong support of one scenario over the other. At issue in this case is OCME's determination that the DNA mixture found on the sweatshirt was 5640 times more likely to have come from defendant and two other unknown individuals rather than from three unknown individuals.
"The long-recognized rule of Frye v. United States . . . is that expert testimony based on scientific principles or procedures is admissible but only after a principle or procedure has gained general acceptance' in its specified field." Wesley, 83 NY2d 417, 422 (1994), quoting Frye v. U.S., 293 F. 1013, 1014 (C.A.D.C. 1923). The Frye test poses the "elemental question of whether the accepted techniques, when properly performed, generate results accepted as reliable within the scientific community generally." People v. Wesley, 83 NY2d at 422. "The Frye inquiry is separate and distinct from the admissibility question applied to all evidence — whether there is a proper foundation — to determine whether the accepted methods were appropriately employed in a particular case." Parker v. Mobil Oil Corp., 7 NY3d 434, 447 (2006). The Court of Appeals has recognized that "the test is not whether a particular procedure is unanimously endorsed by the scientific community, but whether it is generally acceptable as reliable." People v. Middleton, 54 NY2d 42, 49 (1981); see also People v. Wesley, 83 NY2d at 423. A court must conduct a Frye hearing only if the expert testimony involves novel or experimental matters. See People v. Byrd, 51 AD3d 267, 274 (1st Dept. 2008). "[A]s long as the technique itself is generally accepted, a Frye hearing is not necessary even if the application of the technique is unique or modified." People v. Styles, 40 Misc 3d 1205(A), 2013 WL 3286166 (Sup. Ct. Kings County 2013).
The court finds that defendant is not entitled to a Frye hearing because the FST is not new, novel, or experimental. See People v. Belle, 47 Misc 3d 1218(A), 2015 WL 2131497 (Sup. Ct. Bronx County 2015). FST is a software program that uses well-recognized and generally-accepted mathematical principles to compute the likelihood that the DNA of a known individual is included within a sample of DNA containing multiple contributors. Id. Before FST or any other similar software programs were developed, forensic scientists knew how to analyze complex DNA mixtures, but, as a practical matter it could not be done without a computer software program. See People v. Rodriguez, Ind. 5471/09 (Supreme Court, New York County 2013)(Carruthers, J.). FST is merely the OCME's way of computerizing the analytical process in order to make it feasible. Id.
What defendant, and others like him, challenge is the assumptions regarding allele drop-in and drop-out rates made by OCME, which make FST different than other software programs that state the likelihood ratios between the probability that a defendant is a contributor to a mixture and that he is not. However, those matters are relevant to trial foundation or weight of the evidence, "both matters not properly addressed in the pretrial Frye proceeding." People v. Wesley, 83 NY2d at 426 ("defendant's challenges to the population studies relied on by Lifecodes to estimate the probability of a coincidental match go not to admissibility, but to the weight of the evidence, which should be left to the trier of fact"). Every one of defendant's objections to the FST — including the myriad invalid assumptions upon which defendant claims FST is based — could be raised during the cross-examination of the People's expert at trial and the direct testimony of defendant's own expert. See People v. Debraux, 50 Misc 3d 247 (Sup. Ct. New York County 2015); People v. Belle, 2015 WL 2131497 at *6. There are many other software packages, other than FST, that state likelihood ratios regarding the probability that a defendant is a contributor to a DNA mixture and the likelihood that he is not. A defense expert could avail himself of any one of those in order to provide the jury with an alternative likelihood ratio.
Moreover, in 2012, the use of FST was approved by the National Forensic Science Technology Center (NFSTC), an organization founded in 1995 by members of the American Society of Crime Laboratory Directors, which conducts audits of public laboratories conducting forensic DNA testing, in order to meet its stated purpose of "elevating the quality and consistency of forensic services in [the] nation's crime laboratories." People v. Rodriguez, Ind. 5471/09 at 37. In addition to approving FST, the NFSTC also indicated that FST was "not novel and that it met the requirements of a software modification." Id. at 38.
While the court in People v. Collins, 49 Misc 3d 595 (Sup. Ct. Kings County 2015), recently held, after a lengthy Frye hearing, that FST was not generally accepted in the relevant scientific community, the vast majority of the courts of this state that have considered the admissibility of FST have concluded that the techniques used to develop FST are not new, and instead are based on well-established mathematical and statistical principles. See, e.g., People v. Pelt, Ind. 2607/13 (Sup. Ct. Queens County 2015)(Aloise, J.); People v. Julio Lebron, Ind. 791/14 and 1733/14 (Sup. Ct. Queens County 2015)(Lewis, J.); People v. Lyons, Ind. 1433/13 (Sup. Ct. Queens County 2015)(B. Schwartz, J.); People v. Belle, 2015 WL 2131497; People v. Styles, 40 Misc 3d 1205(A)(Sup. Ct. Kings County 2013); People v. Caballero, Ind. N10278/11 (Sup. Ct. Queens County 2013)(Knopf, J.); People v. Brissett, Ind. 847/10 (Sup. Ct. Bronx County 2013)(Webber, J.); People v. Garcia, Ind. 2650/09 (Sup. Ct. Bronx County 2013)(Iacovetta, J.).
Even if FST were new, novel, or experimental, there would be no need for this court to hold a Frye hearing in this case to determine whether FST is generally accepted in the relevant scientific community. Two courts — one in Kings County and one in New York County — have held Frye hearings on FST, and it is not necessary for this court to duplicate those efforts. General acceptance of a scientific principle or procedure may be shown through legal writings and judicial opinions. See Matter of Lahey v. Kelly, 71 NY2d 135, 144 (1987). Rulings by other judges finding a technique generally acceptable can obviate the need for a Frye hearing. See People v. LeGrand, 8 NY3d 449, 458 (2007).
The court has reviewed the decision in Rodriguez, which determined that FST was generally accepted in the relevant scientific community, and the decision in Collins, which determined that it was not. It has also reviewed portions of the hearing testimony in Collins. Ultimately this court comes to the same conclusion reached in Rodriguez and the cases cited above.
Defendant contends that "significant — potentially decision-altering — information emerged" after the Rodriguez decision was issued (Martinez affirmation at ¶35); namely that the New York State Inspector General issued a report critical of Dr. Theresa Caragine, one of the chief developers of FST, and that Drs. Bruce Budowle and Ranajit Chakraborty testified in the Collins Frye hearing that FST was not generally accepted in the relevant scientific community. However, none of the allegations against Dr. Caragine related to the development of the FST system. Moreover, the Inspector General's final report was issued in December 2013, well before the Collins court issued its written decision in that case, and the court specifically stated that it did not find it relevant to Dr. Caragine's testimony at its Frye hearing that she was obligated to leave the OCME "because she resolved a dispute over analysis in a DNA case in a manner inconsistent with OCME protocols about how analyst disagreements should be handled." See People v. Collins, 49 Misc 3d at 607, n.5. Finally, for reasons stated below, this court does not fully credit the testimony of Drs. Chakraborty and Budowle.
The Frye standard was developed to keep " junk science' or novel theories,'" from juries. See Cornell v. 360 West 51st Street Realty, L.L.C., 22 NY3d 762 (2014)(Pigott, J. dissenting). But FST is neither. "OCME conducted exhaustive validation procedures before adopting FST." People v. Garcia, 39 Misc 3d 482 (Sup. Ct. Bronx County 2013); People v. Belle, 2015 WL 2131497 *4. FST was reviewed and unanimously approved for use by the New York State Forensic Science Commission (FSC) and the DNA Subcommittee of the FSC. See People v. Rodriguez, Ind. 5471/09 at 31. The FSC, which was established by Article 49-B, Section 995-a of the Executive Law, includes among its members the chair of the New York State Crime Laboratory Advisory Committee, the director of a forensic laboratory in New York state, the director of the Office of Forensic Services within the Division of Criminal Justice Services, and two scientists experienced in the areas of laboratory standards or quality assurance regulation and monitoring. Executive Law § 995-a. It is tasked with developing "minimum standards and a program of accreditation for all forensic laboratories in New York State." Executive Law § 995-b (1). The subcommittee on forensic DNA laboratories and forensic DNA testing — which includes among its members a representative of the discipline of laboratory standards and quality assurance regulation and monitoring, a forensic scientist, and a representative of the discipline of forensic science — assesses and evaluates "all DNA methodologies proposed to be used for forensic analysis, . . . and make[s] binding recommendations for adoption by the commission addressing minimum scientific standards to be utilized in conducting forensic DNA analysis including, but not limited to, the examination of specimens, population studies and methods employed to determine probabilities and interpret test results." Executive Law § 995-b (13)(a), (b). Even the court in Collins acknowledged that the members of the subcommittee were "world-class scientists in various disciplines relevant to DNA analysis." See People v. Collins, 49 Misc 3d at 612.
In testimony during the Frye hearing held in Rodriguez, John Ballantyne, Ph.D. — who was the chairperson of the DNA Subcommittee of the New York State Forensic Science Commission when it recommended to the full commission that FST be approved for use in casework in New York State — described how the subcommittee reached its decision. During 2009 and 2010, forensic scientists from the OCME made four presentations on FST to the subcommittee, providing detailed information about FST, the manner in which the software accommodates drop-in and drop-out rates, and the extensive validation studies conducted, which Dr. Ballantine described as, "extremely well-documented," and "probably more thorough and better documented than the vast majority of validation exercises [he had] seen." People v. Rodriguez, Ind. 5471/09 at 32. And, as the Rodriguez court noted, "it would be bizarre indeed for a body of such highly accomplished forensic scientists, charged by law with this solemn duty, to recommend [a] software program for use in DNA analysis unless confident that it was firmly based upon principles and methodology accepted as reliable by colleagues in the field." Id. at 35.
Hinda Haned, Ph.D., testified at the Frye hearings in both Rodriguez and Collins. She was a member of the Netherlands Forensic Institute, "one of the principal laboratories in Europe providing forensic DNA analyses for the International Criminal Court in The Hague as well as for authorities in several European nations." Rodriguez, Ind. 5471/09 at 22. In her work, Dr. Haned utilized a program known as Forensim that calculated likelihood ratios that accounted for the probability of drop-in and drop-out in analyzing complex DNA mixtures. Rodriguez, Ind. 5471/09 at 36. Dr. Haned had attended scientific meetings where FST was discussed and had seen a demonstration of the software (Haned testimony at Collins Frye hearing at 30). At one of the workshops she attended, Dr. Haned ran a number of cases through both her program and FST and ended up with similar likelihood ratios (Id. at 35). According to Dr. Haned, several generally-accepted and reliable methods exist for determining drop-in and drop-out rates, including FST (Id. at 32). In fact, at the time of her testimony in the Collins hearing, Dr. Haned was investigating how to determine drop-in and drop-out rates in the same way that FST does, because FST uses more precise drop-out rates and generates more conservative results than her own program (Id. at 33, 43). This court agrees with the Rodriguez court that Dr. Haned — who lectures and publishes extensively in the field — is in "a remarkably authoritative position to provide an accurate assessment of FST's place in the forensic science community." Rodriguez, Ind. 5471/09 at 36. Therefore, this court, like the Rodriguez court, gives great weight to Dr. Haned's opinion that FST and its methodologies are accepted within the forensic science community. Id.
While this court does not deem dispositive that FST has been the subject of several peer-reviewed articles, it does agree with the Rodriguez court that it is notable that these articles and other professional presentations regarding FST have not resulted in a "groundswell of opposition to FST." Id. at 40.
The court in Collins acknowledged that in assessing the admissibility of evidence under Frye, judges should count scientists' votes, rather than attempt to verify the soundness of the scientific conclusion being assessed. See People v. Collins, 49 Misc 3d at 603. Then the Collins court counted the votes differently than did the Rodriguez court. In doing its own counting, this court finds the testimony of the defense witnesses somewhat less persuasive or independent than the Collins court found them. The defense in the Collins hearing called several experts who testified about their concerns regarding how FST handles allelic drop-in and drop-out and that FST is not generally accepted by the relevant scientific community. However, three of the witnesses — Drs. Bruce Budowle, Ranajit Chakraborty, and Angela Van Daal — all either worked for or were affiliated with the University of North Texas at the time of their testimony. Dr. Chakroborty was even recruited to the University by Dr. Budowle. (Chakraborty testimony at Collins Frye hearing at 1213-15). This casts some doubt about his independence and objectivity. Similarly, Dr. Van Daal, who is an external supervisor of several Ph.D. students at the University, had collaborated on several papers with Dr. Budowle ( Id. at 1215-16). Also relevant to a discussion of Dr. Chakraborty's opinion is that he was on the DNA Subcommittee of the FSC that approved FST for use in New York, and only later changed his mind about the validity of the program ( Id. at 1101-02, 1121-24). Another defense witness, Dr. Heather Coyle, was a defense consultant who reviewed only three of forty-two binders of FST validation materials and previously testified in a court case about her concerns about FST without having reviewed any of OCME's validation materials (Coyle testimony at Collins Frye hearing at 463, 467-68, 474). Given what this court perceives to be a possible lack of objectivity guiding the testimony of several of the defense experts in Collins, this court is satisfied that FST is generally accepted in the relevant scientific community.
While on the subcommittee, Dr. Chakraborty also voted to approve the use of low copy number methodology and true allele, and by the time of his testimony at the Collins hearing, had changed his mind about the scientific validity of both of those as well.
It bears noting too that the court in Collins was untroubled by the fact that the same entity seeking to preclude FST evidence in that case has relied on it in other cases when it has suited them. This court is not so dismissive of this contradiction. While it is certainly permissible to take contradictory positions in the defense of different clients, doing so here affects the credibility of the arguments put forth by counsel. It would be one thing for the defense to argue to the jury that the methods used by the OCME were faulty, or that its results were wrong in a particular case, while seeking to admit similar evidence in another. But to argue in one case that the OCME's method of determining a likelihood ratio is not generally accepted while arguing in other cases that it is, is an entirely different matter. The question the court is being asked to consider here is whether or not FST is generally accepted within the relevant scientific community. The answer to that question is either yes or no. It cannot be yes if the results benefit the defense but no if they benefit the prosecution.
For all of the above reasons, this court agrees with the Rodriguez court and others (see, e.g. People v. Debraux, 50 Misc 3d 247; People v. Belle, 47 Misc 3d 1218(A); People v. Brissett, Ind. 847/10), holding that FST is generally accepted within the forensic scientific community.
The court also agrees with the Rodriquez court that it is of no moment, for purposes of Frye, that the FST software is not "open source." Rodriguez, Ind. 5471/09 at 46. Frye requires only that the scientific principle or discovery has achieved general acceptance, and the FST had done that.
The Results Generated by the FST in this Case are Relevant
Having determined that the FST results are admissible under Frye, the court also finds that they are relevant. "[R]elevant evidence is admissible at trial unless [its] admission violates come exclusionary rule." People v Alvino, 71 NY2d 233 (1987). "As a general rule, evidence is relevant if it tends to prove the existence or non-existence of a material fact, i.e., a fact directly at issue in the case." People v. Primo, 96 NY2d 351, 355 (2001). Relevant evidence that is otherwise admissible may be excluded by a court "if its probative value is outweighed by the prospect of trial delay, undue prejudice to the opposing party, confusing the issues or misleading the jury." Id.
The evidence generated by the FST in this case is certainly probative. Rather than merely presenting evidence that a defendant could be a contributor or could not be excluded as a contributor to a DNA mixture, using the FST the People are able to provide a quantitative statistical value to the result of the DNA testing. More specificity regarding the likelihood that a defendant contributed DNA to a mixture contained in a sample is highly probative evidence. Moreover, in this case the probative value is not outweighed by any of the factors laid out in Primo. The inclusion of the evidence will not unduly delay the proceedings, as the People will already be presenting expert testimony regarding other DNA evidence, and the evidence will not confuse the issues or mislead the jury because it will be presented by qualified experts able to explain to lay people the methodologies involved. Defendant may call his own experts to rebut the evidence or explain its limitations. Finally, although the evidence is prejudicial to defendant, evidence of guilt "is expected and supposed to be so. . . . Merely invoking the word prejudice' does not, in and of itself, preclude admission of relevant evidence." People v. Colavito, 87 NY2d 423, 429 (1996). Defendant is not Entitled to Discovery of the FST Source Code.
Defendant's request for an order requiring the People to provide him with the FST source code is denied. "Source codes are the computer instructions followed by a computing device in processing information." People v. Robinson, 53 AD3d 63, 65 (2d Dept. 2008). A computer source code is a written document within the meaning of section 240.20(1)(c) of the Criminal Procedure Law. Id. at 68. However, the discovery statute does not require the People to obtain materials from sources beyond their control, "such as those in possession of the medical examiner." Id. at 73. Here, the People have never possessed the source code. The OCME is an independent agency not under the control of the District Attorney's Office, (see People v. Washington, 86 NY2d 189 [1995]), and the source code is proprietary software copyrighted by the city of New York. People v. Pelt, Ind. 2607/23. Additionally, defendant's contention that he needs the source code in order to fully analyze how the FST works is unconvincing. See State v. Wakefield, 47 Misc 3d 850 (Sup. Ct., Schenectady County 2015). The scientists who testified for the defense at both the Collins and Rodriguez Frye hearings had no trouble detailing the perceived shortcomings of the program without having access to the code. The court finds no reason to believe that defendant's right to cross-examine witnesses or present a defense will be curtailed by the court's refusal to order the People to obtain and disclose OCME proprietary software. Defendant is not Entitled to Discovery of the OCME's Raw Data.
Defendant's request for an order compelling discovery of the electronic raw data relating to the DNA tests conducted in connection with this case is denied for the same reason. Many courts of concurrent jurisdiction faced with discovery motions of this sort have determined that the material sought by defendant here — the OCME's electronic raw data — is not a written report or document, and therefore is not discoverable under section 240.20 of the Criminal Procedure Law. See, e.g., People v. Tsintzelis, Ind. 821/14 (Sup. Ct. Queens County 2014)(Hirsch, J.); People v. Cadlett, Ind. 2376/13 (Sup. Ct. Queens County 2014)(Margulis, J.); People v. Caballero, Ind. 10278/11 (Sup. Ct. Queens County 2012)(Knopf, J.); People v. Beckham, Ind. 1827/09 (Sup. Ct. Queens County 2011)(Griffin, J.); People v. Heyward, Ind. 4714/09 (Sup. Ct. New York County 2010)(Zweibel, J.). However, several Bronx County courts have deemed this type of material discoverable. See, e.g., People v. Franco, Ind. 3760/13 (Sup. Ct. Bronx County 2015)(Clancy, J.); People v. DeJesus, Ind. 3834/13 (Sup. Ct. Bronx County 2015)(Clancy, J.); People v. Grant, Ind. 604/13 (Sup. Ct. Bronx. County 2014)(Fabrizio, J.); People v. Givens, Ind. 348/12 (Sup. Ct. Bronx County 2014)(Webber, J.). This court need not decide that issue, however, because — like the FST source code — the raw data sought by defendant is not in the People's possession and therefore is not discoverable on that basis. See Robinson, 53 AD3d 63.
Defendant's request, in the alternative, for leave to file and serve a subpoena duces tecum on the OCME seeking production of their files is not properly before this court. Leave to file a motion for a court-ordered subpoena is not required. If defendant wishes to subpoena records for trial he should follow the procedures set forth in C.P.L. § 610.20 and C.P.L.R. § 2307, and give notice to the People and the OCME.
This constitutes the decision and order of the court.
The Clerk of the court is directed to forward a copy of this decision and order to the attorney for defendant and to the district attorney. Dated: Kew Gardens, New York January 12, 2016 ______________________________________ Robert A. Schwartz Acting Justice, Supreme Court