UNITED STATES of America, Plaintiff,
v.
Manuel CORONADO-CERVANTES, Defendant.
Criminal No. 94-CR-235 MV.
United States District Court, D. New Mexico.
January 9, 1996.*498 Kelly H. Burnham, Asst. U.S. Atty., Las Cruces, New Mexico, for plaintiff.
Kurt J. Mayer, Asst. Federal Public Defender, Las Cruces, New Mexico, for defendant.
MEMORANDUM OPINION AND ORDER
VAZQUEZ, District Judge.
This matter is before the Court for a determination as to the admissibility at trial of DNA (deoxyribonucleic acid) evidence pursuant to Fed.R.Evid. 702. The United States submitted its Notice of Intent to Use DNA evidence on October 13, 1994. The government filed a pre-hearing memorandum in support of the admissibility of DNA identification evidence on May 17, 1995. Defendant filed a preliminary response on June 6, 1995. The Court conducted hearings on June 6, 1995, July 7, 1995, and August 14 and 15, 1995. Testimony and affidavits were submitted by experts on behalf of both parties. The government and the Defendant filed post-hearing briefs on September 20, 1995.
FACTUAL BACKGROUND
Defendant Manuel Cervantes was indicted on April 20, 1994, under 18 U.S.C. §§ 1153, 2241(a) and 2245(2)(A) and (B)[1] for aggravated sexual abuse. The crime involved an alleged sexual assault that occurred on February 22, 1994, on the Mescalero Apache Indian Reservation in New Mexico. The alleged victim, Virginia Gaines, a 75-year-old member of the Mescalero Apache Tribe, reported on February 24, 1994, that she had been sexually assaulted by her son, the Defendant in this case. A tribal police officer then went with Mrs. Gaines to her house and collected as evidence, her nightgown, camisole and slip, as well as a mattress cover, sheet and blanket from her bed. These items were submitted to the Federal Bureau of Investigation (FBI) laboratory on April 25, 1994.
Preliminary tests by FBI Agent Audrey Lynch revealed the presence of semen on the nightgown, slip and sheet. A sample of Defendant's blood was also submitted at that time. Further evaluation and testing of this evidence revealed that DNA profiles developed from the semen found on those items matched those of Defendant's blood.
Special Agent Lynch computed the likelihood that someone other than the Defendant could have been the source of the evidentiary materials as 1 in 750,000 in Caucasians, 1 in 200,000 in Blacks, 1 in 45,000 in Hispanics and 1 in 10,000 in American Indians.[2] June 6, 1995, Hrg.Tr. at 63.
In a recent comprehensive and well-written opinion, fellow District Court Judge Santiago Campos (D.N.M.) detailed the science and technology of DNA evidence using the RFLP technique, explained the FBI's use of the fixed-fin method of developing population databases and its use of the product rule to estimate the statistical significance of a match of DNA patterns in various population databases and applied the standard enunciated in Daubert v. Merrell Dow Pharmaceuticals, *499 Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993) for determining the admissibility of the DNA evidence. United States v. Peters, Doc. No. 91-CR-395 (D.N.M., Sept. 7, 1995). The Court adopts the extensive explanation in Peters of the underlying scientific procedures at issue in this case and will refer to the Peters opinion throughout its ruling. The Court does, however, independently apply Daubert to the facts before it.
In this case, as in Peters, a Native American defendant stands accused of a sexual crime occurring on an Indian reservation and the validity of the FBI's frequency computation is challenged. Like the defendant in Peters, Defendant Cervantes argues that the FBI lacks adequate Native American databases from which to calculate the expected frequency of his DNA pattern.
Under Daubert, the admissibility of scientific evidence is a preliminary question to be determined by the Court under Fed. R.Evid. 104(a). 509 U.S. at ___, 113 S.Ct. at 2796. The government bears the burden of proof by a preponderance of the evidence. Id. at ___ n. 10, 113 S. Ct. at 2796 n. 10.
The government introduced the expert testimony of Dr. Harold A. Deadman in forensic DNA analysis; Dr. Ranajit Chakraborty, in the area of human population genetics; and FBI Special Agent Audrey Lynch in DNA profiling.[3] The defense presented expert testimony of Dr. Diane K.J. Lavett in molecular biology and Dr. Jeffrey C. Long in anthropology and human genetics.[4]
There are two distinct processes involved in DNA profiling: (1) the Restriction Fragment Length Polymorphism (RFLP) analysis, which is the process by which a laboratory technician analyzes a DNA sample collected at the crime scene and compares its DNA profile with that of the suspect's; and (2) the methodology by which experts estimate the statistical probability of a coincidental match.
This circuit, interpreting Daubert, enunciated a two-prong test for determining the admissibility of scientific evidence that considers first whether the reasoning or methodology are scientifically valid, and second, whether that reasoning or methodology properly can be applied to the facts in issue, that is, whether it is relevant and of assistance in determining a fact in issue. United States v. Davis, 40 F.3d 1069, 1074 (10th Cir.1994), cert. denied, ___ U.S. ___, 115 S. Ct. 1806, 131 L. Ed. 2d 732 (1995) (citing Daubert, 509 U.S. at ___, 113 S.Ct. at 2796). The Court finds that this test must be independently applied to each of the two processes involved in DNA profiling.
(1) RFLP technique.
In pre- and post-hearing briefs filed with the Court, Mr. Cervantes does not raise any specific challenges to the underlying scientific validity of the RFLP analysis.[5] Thus, the Court adopts the analysis and conclusion in Peters and finds that the reasoning and methodology involved in the RFLP technique are scientifically valid and reliable, satisfying Daubert's first prong. See Peters slip op. at 47; See also, United States v. Jakobetz, 955 F.2d 786, 800 (2d Cir.1992); United States v. Martinez, 3 F.3d 1191 (8th Cir.1993) (courts may take judicial notice of the reliability of the underlying theories and techniques involved in DNA profiling using the RFLP process).
Applying Daubert's second prong to the RFLP process, the Court must determine whether the reasoning or methodology properly can be applied to the facts at issue in this case. The inquiry is one of relevance. An obvious fact in issue in this case will be whether the Defendant raped his mother. The Court finds that the DNA evidence will *500 assist the trier of fact and is relevant to this inquiry in that it has "a tendency" to show that the Defendant may be the source of the sperm found at the crime scene.
This circuit has intimated that under Daubert's "relevance" prong, it may be necessary to inquire into whether standard protocol, in the DNA context, was followed in a particular case. United States v. Davis, 40 F.3d 1069, 1074-75 (10th Cir.1994). The Davis court, having found that protocol was followed in that case, expressly declined to address the split in the circuits over whether compliance with protocol is an issue of admissibility or weight. Id. However, the court opined that the examination of protocol should properly fall under Daubert's second prong, explaining "if the offering party does not follow protocol, the scientific evidence may not be relevant under Daubert's second prong because improperly applied science cannot assist the trier of fact." Davis, 40 F.3d at 1074 n. 6.
In the absence of clear directive from the Tenth Circuit, this Court finds that under Daubert's second "relevance" prong, compliance with standard protocol in applying the RFLP technique is essential and goes to admissibility, rather than merely to the weight of DNA evidence as urged by the government. This is so because failure to follow standard protocol in applying the RFLP technique may yield an unacceptably high risk of false positive error. See United States v. Martinez, 3 F.3d 1191 (8th Cir. 1993); see also United States v. Galbreth, 908 F. Supp. 877 (1995) (Daubert analysis applied to polygraph evidence).
In any event, Defendant does not specifically challenge the FBI's compliance with standard protocol in this case and testimony revealed that standard protocol in the RFLP process was followed here. Agent Lynch testified on behalf of the government about the FBI's protocol for DNA testing, the quality control techniques used to ensure accurate results, the specific analysis performed on the semen samples obtained in this case from the victim's clothing, and the test samples taken from the Defendant. Thus, the Court finds the RFLP process satisfies both prongs of Daubert, as it is both scientifically valid and relevant to the facts at issue in this case.
(2) Statistical probability of coincidental match.
When the laboratory work is completed and a DNA profile from a suspect has been declared to match the DNA profile of the perpetrator, it means the suspect cannot be excluded as a possible contributor of the DNA found at the crime scene. Statistics must then be generated to give significance to the match by demonstrating how probable it is that a random match could occur, i.e., that the suspect's DNA profile and the perpetrator's DNA profile would match if they were not the same person.
The FBI develops population databases using "fixed bins" which is a procedure the FBI uses to statistically analyze the data. See Peters, slip op. at 61-64 for detailed explanation of the FBI's fixed-bin methodology. The FBI utilizes what is known as the product rule in estimating the frequencies with which a coincidental match may occur. The validity of the use of the product rule in the probability calculation depends on the independence of the bands whose frequencies are being multiplied. See Peters, at 23-24.
The Defendant's challenges to the DNA evidence in this case center on this second part of the DNA process where experts estimate the statistical probability of a coincidental match. The government seeks to introduce Agent Lynch's statistical probability calculations generated in this case from the FBI's Caucasian, Black, Hispanic and American Indian population databases.
Defendant urges that the FBI's methods for calculating population frequencies for Native Americans fail to meet Daubert requirements. He maintains the FBI relies on an inadequate and unreliable database, which fails to separate individual Indian populations, is not suited to Southwest Indians, and fails to produce conservative results.
The FBI's Native American database consists of approximately 200 individuals, approximately 100 of whom are Sioux and the other 100 are from various tribes. June 6, *501 1995, Hrg.Tr. at 56. Additional precautions are taken by the FBI when using its Native American database due to its relatively small size, in order to further ensure very conservative results. Peters, slip op. at 63-64; June 6, 1995, Hrg.Tr. at 60, ln. 2-8.
Defense witness Dr. Long testified that frequency calculations will be underestimated if an incorrect data base is used because certain sub-populations have higher frequencies of DNA. July 7, 1995, Hrg.Tr. at 193-201.[6]
Dr. Chakraborty has tested and rejected this concern of critics like Dr. Long, that intertribal DNA differences are possibly much larger in Indian communities, which are small, compared to differences among larger subgroups of Caucasians or Blacks. June 6, 1995, Hrg.Tr. at 128. While acknowledging the existence of intertribal DNA differences, he determined the differences "are at best equal to that of the ethnic populations within a broad racial group" and that the differences "do not result into a wrong forensic inference." June 6, 1995, Hrg.Tr. at 129 & 140. Dr. Chakraborty testified that the FBI's frequency computation is valid to provide a conservative estimate. June 6, 1995, Hrg.Tr. at 131-32.
The Court finds the FBI's method of frequency calculations satisfies Daubert's first prong, as it is based upon scientific knowledge and is grounded in the methods and procedures of science. Moreover, the fixed-bin method produces conservative results in that it overestimates the actual frequency of alleles in a population. See June 6, 1995, Hrg.Tr. at 56, ln. 22-25. The methodology sufficiently compensates for any substructuring that may exist in its databases.
In applying Daubert's second "relevance" prong to the probability calculations, the Court finds that it is appropriate to present to the jury only frequency calculations derived from the Native American database. In this case, the victim has accused her son, a Native American, of committing the rape. Thus, the issue before the Court in this case is whether the son committed the crime. The statistical probability calculations involving the Native American database will assist the jury in making this determination and are, therefore, relevant to the facts at issue in this case. There is no need to examine databases of other various populations because the issue is limited to whether a certain identified Native American individual could have been the perpetrator and, if so, the likelihood that he was the perpetrator. The calculation of statistical probabilities of other populations is irrelevant.
Thus, while the Court rejects the government's assertion that frequency calculations go to weight rather than to admissibility, the Court finds the FBI's probability calculations are sufficiently conservative to be scientifically valid and calculations taken from the FBI's Native American database are relevant to the facts of this case and admissible under Daubert.
In addition, Dr. Chakraborty testified he reviewed the case notes, copies of autoradiographs, notes on the frequency computations and indications of the databases used in frequency computations in this particular case, and concluded the FBI followed standard protocol in this case. June 6, 1995, Hrg.Tr. at 134-36. The Court was persuaded that protocol was followed in this second process of DNA profiling dealing with statistical calculations.
Rule 403 Balancing
Defendant urges that the probative value of the DNA evidence is substantially outweighed by the danger of unfair prejudice. He insists that because he and his mother shared the same house, there are plausible explanations for the alleged presence of semen on her bed and clothing, which render the DNA evidence irrelevant. In addition, the victim has recanted her story on one occasion, stating her son did not rape her and she made up the story because she was mad at him. Defendant also points to this recantation as rendering the DNA evidence irrelevant.
*502 To the contrary, given the possible innocuous explanations for the presence of semen at the crime scene and the recantation by the victim, the DNA evidence is not unduly prejudicial. It is relevant to the extent that it has "a tendency" to show that defendant may be the source of the DNA found at the crime scene, consistent with the victim's original allegation. The jury, charged with making the necessary credibility determinations, will properly resolve these factual issues. Under Rule 403, the existence of Defendant's alternative, innocuous and reasonable explanations mitigate in favor of admitting the evidence.
CONCLUSION
As urged by Defendant, this Court has considered, in examining the validity of the forensic application of scientific knowledge to the facts of this case, the size of the underlying database; conservativeness of the probabilities calculated; the application of statistical techniques; and the assumptions underlying the product rule and the statistical techniques. See Def.'s post-hrg. br. at 7.
From its consideration the Court finds the DNA evidence obtained in this case is based on scientific knowledge, as it is grounded in the methods and procedures of science. Daubert, 509 U.S. at ___, 113 S.Ct. at 2795. Further, it is relevant to the extent that it has a "tendency" to show that the Defendant may be the source of the DNA found at the crime scene.
While the general concerns raised by Defendant regarding the statistical probability calculations of Native Americans are legitimate, the Court was persuaded that the FBI's estimates of DNA print frequencies among Native Americans are sufficiently conservative to overcompensate for any "substructuring" that may exist among ethnic populations and, therefore, finds that such evidence is based on a scientifically reliable principle, will assist the trier of fact, and will not unduly prejudice the Defendant in this case.
Therefore, the DNA evidence will be admitted in this case along with statistical estimates of coincidental matches occurring in the Native American population only.
NOTES
[1] Now codified as 18 U.S.C. § 2246(2)(A).
[2] Dr. Chakraborty explained that the correct way to state the frequencies in this case would be "the frequency is no more common than 1 in 750,000 Caucasians ..." June 6, 1995, Hrg.Tr. at 138.
[3] The expert qualifications of Drs. Deadman and Chakraborty are laid out in detail in the Peters opinion at pages 26-27 and 29-30 respectively. Ms. Lynch has a bachelor's degree in biology and a master's degree in cell biology. She worked as a scientific researcher for twelve (12) years, then joined the FBI and spent five years in its Serology Identification Unit and five years in the DNA Analysis Unit.
[4] The qualifications of Drs. Lavett and Long are laid out in the Peters opinion at pages 30-32.
[5] Although not addressed in Defendant's briefs, the Court notes that various challenges were made by Dr. Lavett to the RFLP process which the Court considered and found unpersuasive and refuted by the Government's experts.
[6] Similar concerns involving the "isolated" location of the tribe and genetic differences among racial groups were raised and discounted in Peters, where the defendant was a member of the Navajo Nation.