Opinion by
Judge MARQUEZ.
Defendant, Gregory E. Lindsey, appeals the judgment of conviction entered on jury
verdicts finding him guilty of first degree sexual assault, robbery, burglary, habitual burglar, and habitual criminal. We affirm.
On February 16, 1988, during the early morning hours, a woman was sexually assaulted by a man wearing a mask who broke into her townhome in Colorado Springs.
In January 1988, another Colorado Springs woman had been sexually assaulted, and evidence in the form of vaginal swabs and stains on bedclothes were recovered during the investigation of that offense. On May 1, 1988, in the early morning hours, an intruder again broke into the same woman’s house and attempted to sexually assault her. The intruder fled when a neighbor became aware of the attacker’s presence and telephoned the victim. As a result of the May incident, defendant was arrested and the police obtained samples of his blood.
Blood samples from defendant and the two victims, together with vaginal swabs taken from the two victims shortly after the assaults, were sent to Cellmark Diagnostic Corporation. Cellmark subsequently reported that the DNA (deoxyribonucleic acid) from known blood of the defendant matched the DNA “fingerprint” of the samples obtained from the vaginal swabs of the two victims and a semen stain sample from the bedsheet recovered after the January incident.
The basic procedures used in the Cellmark test, which have been described extensively in
Cobey v. State,
80 Md.App. 31, 559 A.2d 391 (1989), are known collectively as Restriction Fragment Length Polymorphism analysis (RFLP Analysis). This analysis involves separate scientific procedures or experiments including 1) extraction of DNA, 2) fragmentation with restriction enzymes, 3) electrophoresis, 4) Southern blotting, 5) hybridization, 6) autoradiography, and 7) interpretation.
Cellmark then calculates the frequency of DNA.
See State v. Pierce,
64 Ohio St.3d 490, 597 N.E.2d 107 (Ohio 1992). The database used in this case ultimately comprised DNA samples from blood taken from approximately 330 black donors at a Detroit blood bank.
The defendant was originally charged in a single information with separate counts involving the two victims. Ultimately, the counts were severed and the defendant was tried separately on each count.
Prior to severance, the court conducted a pre-trial hearing on the admissibility of the DNA fingerprinting pursuant to
Frye v. United States,
293 F. 1013 (D.C.Cir.1923). Under
Frye,
courts are to determine if the scientific basis of an expert opinion is founded on “a well-recognized scientific principle or discovery [which has] gained general acceptance in the particular field in which it belongs.”
Frye v. United States,
293 F. at 1014.
At the
Frye
hearing, in September 1988, the prosecution initially presented two expert witnesses, each of whom generally testified to the scientific acceptance of the theory of DNA testing and of the DNA fingerprinting technique used by Cellmark. Defense counsel advised the court that he was unable to cross-examine the prosecution’s expert witnesses or to present expert testimony.
In a continuance of the
Frye
hearing in January 1989, after severance and during the trial of the May incident, the prosecution presented a third expert, who testified that the database Cellmark used in arriving at its results is considered sufficient and accepted as reliable in the scientific community. He explained that the procedure by which Cell-mark determines the frequency of banding patterns within the general population is generally accepted in the scientific community and that scientists have been using that principle for quite some time.
Defense counsel again informed the court that a decision had been made not to expend scarce resources challenging the DNA evidence in the trial of the May incident. The trial court then ruled that the DNA evidence was admissible.
In the trial of the May incident, defendant was convicted of second degree burglary, first degree attempted sexual assault, and habitual criminality. In that trial, the DNA evidence served only to link defendant to the similar transaction that had occurred in January 1988. That judgment of conviction was
affirmed by this court in
People v. Lindsey,
(Colo.App. 89CA0340, Sept. 27, 1990) (not selected for official publication).
In the present case, involving the February victim, the prosecutor moved the court to rule that the
Frye
hearing held in September 1988 and January 1989 would be dispositive of the issue of the admissibility of the DNA evidence. Defense counsel moved for a further
Frye
hearing, arguing that, at the first trial, he was not provided with funds with which to challenge the DNA evidence and that developments in the interim raised doubts about the reliability of DNA in a forensic setting.
Based upon Colorado precedent, the court’s consideration of the evidence presented at the
Frye
hearing previously, and decisions in New York, Florida, and Maryland, the court found that the scientific basis upon which both the DNA testing occurs and the scientific basis for analyzing the probability of a particular characteristic through the use of statistics are commonly accepted within the scientific community and thus, the
Frye
standard had been met. Accordingly, it denied a further
Frye
hearing. However, the court specifically held that it was not ruling that the application or the use of the techniques in the individual eases of Cellmark was or was not admissible. Instead, it ruled that such issue required a motion
in limine.
Defendant then filed a motion
in limine
to exclude the DNA evidence, and a hearing was held on that motion. Defendant also moved to reconsider the court’s ruling regarding the applicability of the
Frye
case. However, the trial court denied the latter motion.
Defendant presented a number of witnesses who testified both at the motion
in limine
hearing and at trial. These experts raised a number of criticisms concerning Cellmark, including that the database upon which it premised its statistical comparisons was too small to be reliable, that its procedures were flawed and had not been reviewed by panels of independent scientists, that standards for the use of DNA fingerprinting for forensic purposes should be more strict than those Cellmark required, and that there are critical differences between forensic and research laboratory settings.
A prosecution expert testified at the
in limine
hearing that the DNA patterns in the Lindsey and unknown sample were indistinguishable and that he had used Cellmark’s data to make his own computations. He stated that he was absolutely confident that the frequency of this pattern in the population could be no more common than one in twenty-one million.
At the conclusion of the
in limine
hearing, the court first noted that the techniques and theories used by Cellmark in this case are commonly accepted in the scientific community. The court weighed the testimony at the
Frye
hearing and the
in limine
hearing and then applied CRE 702. It found that the witnesses produced by both the prosecution and the defendant were qualified experts, that the testimony would assist the trier of fact in determining a fact in issue, and that the existence of significant disagreement among experts was not a reason to exclude the testimony from the trier of fact. There was no reason, in the trial court’s view, for the court to usurp the role of resolving conflicts of fact and opinion. The court then applied CRE 403 and denied the defendant’s motion
in limine.
At trial, the technician who actually performed the RFLP testing technique on the samples in this case testified as an expert for the prosecution. She stated that she had performed tests on nearly 500 forensic samples and testified concerning Cellmark’s laboratory protocols and procedures when testing forensic samples.
A genetic epidemiologist also testified for the prosecution at the trial. She testified that she had reviewed the calculations by which Cellmark arrived at the figure that the odds of someone besides the defendant having the banding pattern appearing in the known sample and in the forensic sample was one in three hundred forty billion. She felt that Cellmark’s data base was sufficient to calculate this frequency in the general black population.
A third expert testified as a prosecution rebuttal witness concerning the various pro
cedures Cellmark uses in doing its RFLP analysis and discussed the reliability of those procedures.
With the exception of the DNA fingerprints, the evidence against the defendant in this ease was largely circumstantial. The victim was unable to identify the intruder, but reported that he was tall, of large build, with very broad shoulders, and that he identified himself as a “black man.” The defendant, who lived next door to the victim, was a tall black man. The victim testified that the intruder wore a dark snowsuit. The defendant owned a pair of black ski pants.
On appeal, defendant asserts that the forensic DNA identification test performed by Cellmark in this case is not generally accepted as rehable in the scientific community. We conclude that the court did not err in admitting this evidence.
A.
Defendant argues that the issue in this case is not whether the theory of DNA matching is correct or whether DNA testing might someday become reliable, but whether the DNA testing procedures employed by Cellmark in this case are reliable and generally accepted by the- scientific community.
Colorado applied the
Frye
test in
People v. Anderson,
637 P.2d 354 (Colo.1981) in resolving the issue of the admissibility of polygraph tests.
However, in
Campbell v. People,
814 P.2d 1 (Colo.1991), our supreme court stated that the
Frye
standard of general acceptance within any particular scientific field was limited to those situations involving novel scientific devices or processes involving the evaluation of physical evidence. Thus, the court rejected the
Frye
standard for testimony by a defendant’s expert about the reliability of eyewitness identification, instead holding that an analysis under CRE 702 and CRE 403 was appropriate.
In
People v. Fishback,
829 P.2d 489 (Colo.App.1991) (cert. granted May 11, 1992), this court addressed the question whether the
Frye
standard applied to DNA fingerprinting. There, while DNA fingerprint evidence was deemed admissible, the particular DNA laboratory is not identified, and there is no indication in the opinion that the defendant presented expert testimony. Nevertheless, we deem the analysis utilized in
Fishback
to be applicable here.
Although the court in
Fishback
also analyzed the DNA evidence under CRE 702 and found it admissible under that rule, it nevertheless considered the admissibility of the DNA evidence under
Frye
and the three-prong test delineated in
People v. Castro,
144 Misc.2d 956, 545 N.Y.S.2d 985 (N.Y.Sup.Ct.1989). In determining admissibility of DNA evidence, the
Castro
court set forth the following test:
Prong I.
Is there a
theory,
which is generally accepted in the scientific community, which supports the conclusion that DNA forensic testing can produce reliable results?
Prong II.
Are there
techniques
or experiments that currently exist that are capable of producing reliable results in DNA identification and which are generally accepted in the scientific community?
Prong III.
Did the testing laboratory perform the accepted scientific techniques in analyzing the forensic samples in this particular case?
People v. Castro, supra,
144 Misc.2d at 959, 545 N.Y.S.2d at 987 (emphasis added).
The
Castro
court held that the first two prongs of the analysis deal strictly and exclusively with the
Frye
issue. Under prong III of the
Castro
test, a pre-trial hearing is to be held to determine if the testing laboratory performed the accepted scientific techniques yielding sufficiently reliable results to be admissible as a question of fact for the jury.
Assuming that the
Frye
standard and the three-prong test of
Castro
are applicable here, we turn to the question whether the requirements of
Frye
and
Castro
have been met in this case.
At oral argument, defendant acknowledged that it is uncontroverted that RFLP analysis has attained general acceptance in the scientific community for use in research laboratories. However, while conceding that prong I of the
Castro
test has been met in its entire
ty, and prong II has been met, at least insofar as it applies to the laboratory portion of the test, defendant nonetheless disputes whether the forensic application of the statistical portion of the evidence is generally accepted as reliable by the scientific Community-
Defendant disputes prong III in its entirety. He asserts that Cellmark’s forensic DNA test has not gained general acceptance in the scientific community, arguing that initial uncritical acceptance of forensic DNA identification tests in other jurisdictions does not constitute persuasive precedent of general acceptance of Cellmark’s test.
In
People v. Fishback, supra,
the court determined that prong I of the
Castro
test was satisfied. This determination is in accord with a number of states which have adopted the
Castro
test.
See Ex parte Perry,
586 So.2d 242 (Ala.1991);
Hopkins v. State,
579 N.E.2d 1297 (Ind.1991); and
People v. Castro, supra.
In addition, the theory underlying DNA forensic techniques has been accepted in such eases as
People v. Miles,
217 Ill.App.3d 393, 577 N.E.2d 477 (1991);
Smith v. Deppish,
807 P.2d 144 (Kan.1991);
Cobey v. State, supra; State v. Schwartz,
447 N.W.2d 422 (Minn.1989);
State v. Pennington,
327 N.C. 89, 393 S.E.2d 847 (1990);
Spencer v. Commonwealth,
238 Va. 275, 384 S.E.2d 775 (1989),
cert. denied,
493 U.S. 1036, 110 S.Ct. 759, 107 L.Ed.2d 775 (1990). Therefore, we conclude that, as a matter of law, prong I of the
Castro
test has been satisfied for DNA fingerprint evidence.
With respect to prong II, defendant argues that acceptance of RFLP Analysis for research, diagnostics, and paternity testing does not imply acceptance of forensic DNA testing. As regards forensic work, he asserts that evidentiary samples are limited, that greater precision is required to analyze forensic samples, that there is no consensus among the three major forensic laboratories regarding what matching rule is appropriate, that Cellmark has not published its matching criteria, that there are as yet no national standards about what a forensic DNA lab should consist of, and that these differences undermine the reliability of the RFLP process as applied to forensic testing. At oral argument on appeal, defense counsel focused especially on the lack of an adequate population base.
Defendant also argues that the scientific community demands that the reliability of forensic DNA testing be examined by distinguished scientific boards and panels and that those groups have yet to reach a consensus on the issue. He further argues that Cell-mark’s failure to publish its work in peer review journals and its “dismal” performance on proficiency tests have prevented the scientific community from assessing the reliability of the Cellmark test or generally accepting its results.
Defendant also contends that the presence of contaminants in forensic samples and the absence of adequate scientific controls undermine the reliability of Cellmark’s forensic DNA test. In this regard, defendant presented expert testimony that Cellmark’s policy of including ethidium bromide in its gels actually promotes band shifting. Another defense expert testified that either a “monomorphic probe” or a “mixing experiment” would permit a determination whether band shift is responsible for differences between samples, and that Cellmark does not use either a monomorphic probe or mixing experiments. However, one of these experts also testified there was no evidence of contamination or band shifting in this case.
The court in
People v. Fishback, supra,
determined that prong II was satisfied because expert testimony indicated that techniques exist that are capable of producing reliable results in DNA identification which are generally accepted in the scientific community. This holding is again in accord with a number of decided cases,
see Ex parte Perry, supra; People v. Castro, supra; State v. Pennington, supra; Hopkins v. State, supra; Spencer v. Commonwealth, supra;
and is even in accord with cases which have decided against the admission of DNA fingerprinting evidence.
See People v. Castro, supra.
As in
People v. Fishback, supra,
expert testimony was presented here supporting the existence of appropriate techniques under
prong II of the
Castro
test. . Considering the evidentiary support and acceptance in
Fish-back
and other jurisdictions, we are satisfied that prong II requirements have been met. The question, then, is whether in this case prong III of the
Castro
test was satisfied.
B.
• Courts in other jurisdictions have reached varying results with respect to the admissibility of Cellmark laboratory tests. Cell-mark DNA evidence was admitted in
Cobey v. State, supra
(no experts testified for defendant);
State v. Davis,
814 S.W.2d 593 (Mo.1991),
cert. denied,
— U.S. —, 112 S.Ct. 911, 116 L.Ed.2d 812 (1992) (no defense expert);
State v. Pennington, supra
(one defense expert);
People v. Miles, supra
(no defense experts);
Hopkins v. State, supra
(defendant presented experts); and
State v. Pierce, supra.
Conversely, evidence based on Cellmark Laboratory testing was held inadmissible in
People v. Barney,
8 Cal.App. 4th 798, 10 Cal.Rptr.2d 731 (1 Dist.1992);
State v. Schwartz, supra; Commonwealth v. Cumin,
409 Mass. 218, 565 N.E.2d 440 (1991) (defendant presented an expert, but the prosecution presented no expert to support Cellmark’s conclusion); and
Commonwealth v. Lanigan,
413 Mass. 154, 596 N.E.2d 311 (1992).
While reaffirming that the admissibility of novel scientific evidence is determined by application of the
Frye
standard, the court in
State v. Schwartz, supra,
ruled that Cell-mark’s test results were inadmissible. The court there focused on Cellmark’s admission that in a proficiency test it had falsely identified two samples as coming from the same subject, that the director of Cellmark’s research and development laboratory admitted that because Cellmark did not meet all the minimum guidelines, such as formal methodology validation and published results of experimental studies in peer review journals, that the FBI likely would not consider its test results ready for use in court, and that the defense request for more specific information regarding its methodology and population database was denied by Cellmark.
In rejecting the DNA evidence, the court also relied on its prior holding in another case that while expert interpretation of scientific results is not foreclosed, a limitation on the use of population frequency statistics is necessary because of the danger that such evidence will have a potentially exaggerated impact on the trier of fact.
Cellmark evidence was also rejected in
Commonwealth v. Cumin, supra.
However, in that case, the prosecutor presented no expert to support Cellmark’s conclusion. Defendant’s expert testified that Cellmark’s database (200 samples collected at a New York City blood bank) was not adequate for the purpose of producing an estimate of the frequency of finding a particular genotype in the human population. Based on the absence of general acceptance or inherent rationality in the process upon which Cellmark based its conclusions, the court concluded that the admission in evidence of the test results was prejudicial error.
In recent cases in which the defendant presented experts in cases not involving Cell-mark Laboratories, some courts have admitted DNA fingerprinting evidence.
See Spencer v. Commonwealth, supra; Smith v. Deppish, supra.
In
Ex parte Perry, supra,
in which the
Castro
test was used, the court held only that the DNA matching evidence met prongs I and II of the test. The court expressed its concern that the evidence was insufficient to determine whether there was error in the performance or interpretation of the test, that DNA population frequency statistics may create a potentially exaggerated impact on the trier of fact, and remanded the case to the trial court for a hearing to determine whether this prejudicial effect outweighed the probative value of the statistical evidence.
Thus, concerns in recent cases have been directed largely to the performance of techniques in a given case, rather than to the underlying theory, or whether accurate techniques exist; thus, they go to prong III of the
Castro
test.
Here, as in other cases, the question whether the techniques involved met the demands of prong III is the subject of conflicting testimony by expert witnesses.
Central to defendant’s contentions in this case is his assertion that the scientific community does not generally accept Cellmark’s statistical calculations. He contends that Cellmark’s assumption that blacks are a homogeneous, randomly mixing group is unproven and contradicted by empirical evidence and that, if there is substructuring among blacks, Cellmark’s database may be totally invalid for drawing conclusions about the frequency of genetic characteristics of black males. He further contends that defendant’s expert testimony shows that Cell-mark’s database shows gross deviations from Hardy-Weinberg equilibrium. Thus, in defendant’s view, the necessary prerequisites for application of the product rule are not present in this case, and the calculations based upon this database are invalid.
C.
In
People v. Castro, supra,
the court stated that because the testimony of the expert scientists demonstrated conflicting opinions, an issue of fact was created which would have been submitted to the jury. It further stated that: “[Ijssues of fact which arise as a result of the hearing concerning the reliability of any particular test, or the size or ratio of the population frequency, relates to the weight of the evidence and not its admissibility.”
People v. Castro,
144 Misc.2d at 979, 545 N.Y.S.2d at 999.
Similarly in
Hopkins v. State,
579 N.E.2d at 1303, with respect to the third prong, the court stated:
We therefore hold that once the trial court has ruled the witness qualified as a matter of law to give expert testimony regarding DNA analysis, the subsequent evaluation of that evidence goes only to its weight as a matter of fact. Any battle of qualified experts, as in the instant case, or other conflict as to the reliability of evidence is to be resolved by the trier of fact, ... whose finding in this regard will be upheld on review as long as the favorable evidence adequately supports it, as with any sufficiency question.
The Colorado Rules of Evidence provide that all relevant evidence is admissible, except as otherwise provided by the United States or Colorado Constitutions, by the rules themselves, by other rules prescribed by the supreme court, or by the Colorado statutes. CRE 402. Although relevant, evidence may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence. CRE 403.
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise. CRE 702.
See Campbell v. People, supra.
We conclude that the considerations weighed by the courts in
Castro, Hopkins,
and other cases concerning conflicting expert opinions are not unlike those embodied in CRE 702 and CRE 403.
Here, at the conclusion of the motions hearing, the trial court determined that it would base its analysis solely on CRE 702 and CRE 403. We conclude that the evidence was admissible under both
Frye
and CRE 702 and 403 and that the
in limine
hearing served the same purposes as a hearing under prong III of the
Castro
test.
Questions concerning the admissibility of evidence are determined by the trial court. CRE 104(a). Its decision will not be disturbed absent an abuse of discretion.
See People v. Hampton,
746 P.2d 947 (Colo.1987).
The admissibility of DNA evidence in this ease was the subject of contested and conflicting testimony. However, substantial expert testimony supports the trial court’s ruling. Highly respected authorities in the areas of genetics and biology testified to the general acceptance and reliability of DNA fingerprinting within the scientific community, as well as to the existence of extensive peer review of the forensic application of DNA fingerprinting techniques.
The reliability of Cellmark’s specific tests is supported by equally strong evidence. There was expert testimony that the methodology in Cellmark’s laboratory substantially corresponds to that in research laboratories
and that Cellmark actually employs a greater number of checks and procedures for ensuring that each step is performed in the proper manner.
Finally, there was strong evidentiary support for the reliability of the statistical calculations used in this case. In particular, the prosecution’s witness at the motion
in limine
hearing addressed many of the specific contentions raised by the defense. He testified that, in response to these concerns, he performed his own calculations, using very conservative bin sizes, standard deviations, and probabilities, all of which heavily favored the defendant. Based upon these figures, he determined that the frequency of the pattern in the population could be no more common than one in twenty-one million. He also provided testimony refuting defendant’s challenges to Cellmark’s database.
Coneededly, the defendant offered several expert witnesses of his own, each of whom concluded that the DNA evidence was flawed. However, based upon the above authorities, the trial court acted within its prerogative in allowing the jury to consider evidence of Cellmark’s tests. We will not disturb that decision.
Judgment affirmed.
METZGER and REED, JJ., concur.