Steele, J.
A Superior Court jury found the defendant guilty of rape and murder in the first degree of a two-year-old child. Prior to trial, the defendant filed a motion in limine, seeking to prevent the introduction of DNA evidence to establish identity. After a hearing, the motion was denied. During the course of the trial, the defendant moved twice for a required finding of not guilty. Both motions were denied. Following the guilty verdicts and after the discharge of the jury, the defendant filed a Motion After Discharge of the Jury Pursuant to Mass.R.Crim.P. 25(b)(2). This motion was [169]*169initially denied. However, upon further consideration of the admissibility of DNA evidence, as a matter of law, this court now allows the defendant’s post-conviction Rule 25(b)(2) motion, sets aside the guilty verdicts and orders a new trial.
BACKGROUND
On April 15, 1993, a Bristol County grand jury returned indictments against the defendant for the murder and rape by force of a child. During a pre-trial conference, the Commonwealth notified defense counsel and this court that it intended to rely on evidence of a deoxyribonucleic acid (DNA) analysis conducted by the Federal Bureau of Investigation in order to establish the defendant’s identity as the perpetrator.1 The FBI laboratory had compared the DNA taken from a sample of the defendant’s bodily fluids with the DNA extracted from semen recovered on an autopsy swabbing of the child’s throat, and concluded that the DNA “matched.”2 The Commonwealth acknowledged that without the evidence of the DNA comparison, it would not have had sufficient evidence to convict the defendant.
The defendant filed a motion in limine seeking to exclude any use of or reference to the DNA evidence. As grounds for this motion, the defendant argued that the DNA evidence was not admissible under the standard articulated in United States v. Frye, 293 F. 1013 (D.C. Cir. 1923), and Commonwealth v. Curnin, 409 Mass. 218 (1991). Upon defendant’s motion, this court ordered that a voir dire hearing be held to determine whether the evidence of a DNA comparison was generally accepted by the relevant scientific community.3
The issue which emerged during the voir dire hearing, and the only part of the DNA evidence challenged by the defendant under the Frye-Curnin standard, was the statistical method used to calculate the significance of a laboratory “match” of DNA profiles. In other words, the defendant did not object to the use of the RFLP laboratory process or the FBI’s implementation of that process in this case. Rather, the defendant challenged the statistical calculation used to arrive at an estimate, in terms of probability, of how likely it would be that another person’s DNA would “match” the DNA extracted from the autopsy swabs of Hallie Thibault. The defendant argued that the statistical calculation used by the Commonwealth to establish that he was the source of the DNA recovered at the autopsy does not pass the Frye-Curnin test.
The voir dire testimony regarding this statistical calculation can be summarized as follows. The Commonwealth presented the testimony of Supervisory Special Agent John Mertens of the FBI serology and DNA analysis unit. After testifying that he holds a bachelors degree in biology, a masters degree in microbiology, and that he specializes in the serology and DNA laboratory techniques performed on physical evidence sent to the FBI laboratory, Mertens was qualified as an expert in DNA analysis and forensic serology.
Agent Mertens testified that he had performed the RFLP analysis, compared the profile of the defendant’s DNA with the profile of the DNA recovered at the autopsy, and concluded that they “matched.”
To estimate how rare or how common it would be to find another person in the general population with a DNA profile that “matched” the profile of the DNA recovered at the autopsy, Mertens explained that the FBI normally uses a statistical calculation which he referred to as the “fixed bin method.”4 Using this statistical calculation, the “matched” DNA profile in this case was compared to DNA profiles which are stored in genetic population databases. The FBI maintains four databases, Caucasian, Black, Hispanic and Native American.
Mertens testified that the “fixed bin” statistical calculation is generally accepted by the scientific community, that it has been accepted by many courts throughout the country, and that this acceptance is reflected in scientific literature.
Applying this statistical calculation to the present case, Mertens had concluded that the chance of going out into the general population and selecting another individual at random whose DNA “matched” the DNA recovered at the autopsy was one in 800,000 in the Caucasian population, one in 7,000,000 in the Black population, and one in 500,000 in the Hispanic population.5
Mertens testified that in a 1992 report, the National Research Council (NRC)6 had expressed misgivings regarding the statistical calculation used by the FBI. Mertens explained that the NRC was concerned that genetic subpopulations might exist within large genetic database populations, such as those used by the FBI. The NRC had recognized that the existence of these genetic subpopulations might affect the statistical calculation used to determine a DNA “match.”
To address the subpopulation issue, the 1992 NRC Report proposed a new statistical calculation known as the “ceiling” approach. The objective of the ceiling method is to eliminate the effect of genetic sub-populations by mathematically adding a layer of conservancy when calculating the frequency of DNA profiles.
Mertens testified that FBI studies indicate that genetic subpopulations do not exist, and that the FBI stood behind the “fixed bin” statistical calculation. However, in response to the Commonwealth’s request, Mertens also provided a computer generated statistical calculation of the probability of a DNA "match” in this case using the ceiling approach. For purposes of the ceiling approach calculation, Mertens testified that the FBI’s Hispanic database was divided into Southeastern Hispanic and Southwestern Hispanic, because FBI studies had found that a difference in allele frequencies exists between people of Cuban descent and people of Mexican descent.7 Based on the ceiling approach calculation, Mertens testified that the chance of going into the population and selecting another individual with the same [170]*170DNA profile as that recovered at the autopsy would be approximately 1 in 159,000.
The defendant’s witness, Dr. Laurence Mueller, Ph.D., testified at the voir dire hearing as follows. Dr. Mueller testified that he holds a bachelors degree in chemistry, a masters degree in biology, a Ph.d in ecology, and that he had done graduate work in genetics and post-doctoral research in the field of theoretical population genetics. Dr. Mueller testified that he specialized in the fields of population genetics and evolutionary biology. Dr. Mueller explained that his work in the field of population genetics involves studying and quantifying genetic variation in natural populations. Dr. Mueller was qualified as an expert in the field of population genetics.
Dr. Mueller explained that the 1992 NRC Report contained several major permanent and interim recommendations on how to statistically evaluate DNA laboratory evidence.8
Dr. Mueller testified that the ceiling principle recommended by the NRC was designed to account for the possibility that genetic substructuring might occur within the large genetic databases currently used in DNA comparisons. Dr. Mueller stated that his own research confirmed that ethnically diverse genetic subpopulations do exist within the larger genetic databases.
Dr. Mueller also testified that in the course of his work as a population geneticist, he had become aware of a controversy within the scientific community surrounding the proposed ceiling method and the interim ceiling method. Dr. Mueller described in detail how the scientific community was divided into three groups over the soundness of the use of the ceiling principle in forensic applications.
The first group of scientists has criticized the 1992 NRC report as being scientifically regressive and the recommended ceiling principles as not being grounded in sound science. This group of scientists has taken the position that the product rule statistical calculation is satisfactory.
The second group of scientists has taken the position that the ceiling principle will not always result in a conservative statistical calculation, and that in some cases it may result in a DNA profile frequency estimate that is rarer than it should be. This group of scientists is also concerned with the nature of the genetic databases used by the FBI.
The third group of scientists includes those who believe that under the proper circumstances, the ceiling principle can provide the corrective conservative statistical calculation which the NRC proposed.
Dr. Mueller included himself in this third group. He testified that he would support the ceiling principle if it were properly implemented using 15-20 database populations. However, Dr. Mueller questioned the correctness of the interim ceiling method because of the low number of databases currently used. In Dr. Mueller’s opinion, the fact that the NRC had recommended the counting method statistical calculation as an adjunct to the interim ceiling calculation reflected the NRC’s own lack of confidence in the interim ceiling statistical calculation alone.
Dr. Mueller testified that the FBI’s method of statistically calculating the frequency of a laboratory “match” departs significantly from the NRC’s interim ceiling recommendation. Dr. Mueller explained that the FBI does not apply the counting method statistical calculation in conjunction with the interim ceiling method, as recommended by the NRC. Dr. Mueller also stated that the FBI ignores the obvious implication of the NRC’s recommendation that as many databases as possible be used in the interim period. Dr. Mueller explained that, although the 1992 NRC Report recommended use of databases such as Asian and Native American, the FBI had not used its own Native American database in this case. Moreover, Dr. Mueller testified that the FBI fails to make use of at least a dozen available databases, representing well defined subpopulations, which are maintained by laboratories across the country using FBI-compatible DNA analysis techniques. Dr. Mueller further testified that in estimating individual DNA band frequencies, the FBI uses a system known as the fixed bins in situations when the NRC would recommend use of a system referred to as floating bins. Dr. Mueller also testified that the FBI departs from the NRC’s recommendations because, although the FBI conducts internal proficiency monitoring, FBI laboratory error rates are not reported to juries.
In Dr. Mueller’s opinion, the FBI’s use, in this case, of a ceiling principle mathematical correction alone was not consistent with the NRC’s three-part interim ceiling recommendation, and was not generally accepted by the scientific community.
Using databases which included a total of 3,515 DNA profiles, and three variations of the ceiling principle statistical calculation. Dr. Mueller had concluded that the likelihood of finding another individual in the population whose DNA would “match” the DNA recovered at the autopsy was 1 in 136, 1 in 351, or 1 in 4,300.9
In addition to the testimony of Agent Mertens the Commonwealth offered numerous exhibits at the voir dire hearing. Among the Commonwealth’s exhibits is the NRC’s April 1992 Report, “DNA Technology in Forensic Science” (NRC Report).10 The third chapter of the NRC Report is entitled, “DNA Typing: Statistical Basis for Interpretation.” The introduction to this section states that, “Substantial controversy has arisen concerning the methods for estimating the population frequencies of specific DNA typing patterns. Questions have been raised about the adequacy of the population databases on which frequency estimates are based and about the role of racial and ethnic [171]*171origin in frequency estimation.” (NRC Report, p. 3-1.) The Report later explains that “The key question underlying the use of the multiplication [product] rule is whether actual populations have significant substructure for the loci used for forensic typing. This has provoked considerable debate among the population geneticists: some have expressed serious concern about the possibility of significant substructure, . . . and others consider the likely degree of substructure not great enough to affect the calculations significantly . . .” (NRC Report, p. 3-6.) The Report goes on to say, “Although mindful of the controversy, the committee has chosen to assume for the sake of discussion that population substructure may exist and provide a method for estimating population frequencies in a manner that adequately accounts for it.” (NRC Report, p. 3-7.)
At the conclusion of the voir dire hearing, the introduction of the DNA evidence was allowed by this court with grave misgivings. Because this court had not previously been faced with the question of the admissibility of the statistical calculations used to interpret DNA evidence, upon defendant’s request, this court later accepted additional relevant written materials from the defendant. These materials were marked as exhibits.
Among the defendant’s exhibits is a package of letters, memoranda and proposals relevant to a new 1993-1994 NRC project to be entitled, “DNA Forensic Science: An Update.”11 Included in the package is an April 1993 memorandum from Philip Smith, Executive Officer of the National Academy of Sciences, explaining that FBI Director William Sessions had requested that the NRC undertake a new study to consider how the most recent scientific studies affect the statistical procedures recommended by the NRC in its 1992 Report. This memorandum points out that the NRC’s own Committee on National Statistics has criticized the statistical approach recommended in the NRC’s 1992 Report, and that numerous statisticians had requested that the NRC reexamine its 1992 Report.
Also included in this exhibit is the April 16, 1993 letter from FBI Director William Sessions to the National Academy of Sciences President, Dr. Frank Press. The letter begins by stating that, “Last year’s report by the National Research Council (NRC) on DNA Technology in Forensic Science has been the subject of a national debate that has created a climate of confusion for courts considering the admissibility of DNA evidence.” After requesting that the NRC reexamine its 1992 recommendations, the letter explains that the “crisis largely centers around controversy over the report’s recommendation to use a ‘ceiling principle’ in calculating probabilities of DNA matches.”
Also included in this exhibit is a letter dated April 7, 1993 forwarded to Frank Press, Chairman of the NRC, signed by over 100 scientists attending the Second International Symposium on the Forensic Aspects of DNA, held at the FBI Academy. The letter from the group of scientists states that the “statistical and population genetic recommendations of the [NRC] report have not been well received by the scientific community.” The letter goes on to request that the NRC revisit statistical and populations issues surrounding forensic uses of DNA.
The Prospectus for the newly proposed NRC study, “DNA Forensic Science: An Update,” is also included in defendant’s exhibit.12 The Prospectus begins by explaining that, “In 1992 the National Research Council (NRC) published ‘DNA Technology in Forensic Science.’ The report was immediately cited in court cases as the consensus of scientific opinion for determining the admissibility of DNA typing as evidence and for specifying the statistical approaches to be used in interpreting the data from this source of information. Since its publication, the report has been the subject of close attention and debate.”
FINDINGS OF FACT
After reconsidering all of the evidence described and referred to above, I find that the ceiling principle and the interim ceiling principle are not generally accepted within the relevant scientific community. I also make the following findings of fact.
1. FBI Agent Mertens specializes in DNA laboratory techniques, and he is not an expert in the field of population genetics.
2. Mertens’s ability to give an opinion as to the general acceptance of the ceiling principle within the scientific community was limited to his perception of the FBI’s position in the controversy. For this reason, I give little weight to his opinion on this issue.
3. There is a debate in the relevant scientific community over whether ethnically diverse genetic sub-populations exist within the larger “heterogeneous" populations represented in the databases currently used in DNA analysis.
4. Even if, for the sake of argument, it is assumed that these genetic subpopulations do exist, I find that there is disagreement in the relevant scientific community as to whether the subpopulations affect the statistical calculations used to estimate the frequency of a DNA “match.”
5. The FBI’s position on subpopulations as presented by Agent Mertens was internally inconsistent: Agent Mertens testified that FBI studies have found that no subpopulations exist, yet he also testified that the FBI’s Hispanic database is broken down into Southeastern Hispanic and Southwestern Hispanic, due to genetic variations between people of Cuban and Mexican descent. I find that Mertens’s testimony on this issue was not credible.
6. This inconsistency reflects a lack of consensus within the FBI as to the existence and effect of genetic subpopulations on the statistical calculations used to interpret DNA “matches.”
[172]*1727. Dr. Mueller specializes in the study of ethnically diverse subpopulations found in population databases, and he is an expert in the field of population genetics.
8. Dr. Mueller was well informed as to the range of scientific opinion represented in the national debate over the proper statistical calculation to use in reliably determining the frequency of DNA profiles in the general population. I find that Dr. Mueller’s testimony was credible, and I accord his testimony great weight.
9. The April 1992 NRC Report entitled, “DNA Technology in Forensic Science,” is evidence of the national debate over the existence and effect of genetic sub-populations on DNA statistical calculations. The 1992 NRC Report does not represent any consensus in the scientific community relevant to the statistical calculations.
10. The 1992 NRC Report’s recommendation of the use of ceiling and interim ceiling methods further fuelled the controversy concerning the appropriate statistical calculation to be used when interpreting DNA evidence.
11. There is disagreement within the relevant scientific community over whether the ceiling principles provide for a more conservative statistical calculation of DNA profile frequencies than the product rule.
12. Some scientists in the relevant scientific community believe that the ceiling approach may produce less conservative estimates of DNA profile frequencies.
13. The FBI did not apply the three-part interim ceiling method statistical calculation as recommended by the NRC, in this case.
14. There is no support in the relevant scientific community for the FBI’s use of a corrective mathematical ceiling alone.
15 I find that a) the study proposed by the NRC in 1993 entitled, “DNA Forensic Science: An Update,” b) the letter dated April 16, 1993 from FBI Director William Sessions to the NRC, c) the petition with more than 100 signatures of scientist asking the NRC to revisit the issue of DNA statistical calculations, and d) the scientific articles submitted as exhibits in this case clearly establish that there is an ongoing dispute and lack of general consensus regarding the soundness of the ceiling and interim ceiling methods proposed by the NRC in 1992.
16. The state of the scientific discourse is such that currently there is no statistical calculation for interpreting the significance of a laboratory DNA “match” which is generally accepted by the relevant scientific community.
DISCUSSION
Exclusion of DNA Evidence
To decide whether evidence produced by a scientific theory or process is admissible in court, the theory or process must be generally accepted by the community of scientists involved. Frye v. United States, 293 F. 1013 (D.C. Cir. 1923); Commonwealth v. Curnin, 409 Mass. 218, 222 (1991); Commonwealth v. Mendes, 406 Mass. 201, 205 (1989). Evidence of “matching” DNA profiles is inadmissible, if the statistical calculation used to determine the frequency of such a “match” is not generally accepted by the relevant scientific community. Commonwealth v. Daggett, 416 Mass. 347, 350 (1993); Curnin, supra at 222 n.7 (“we would not permit the admission of test results showing a DNA match (a positive result) without telling the jury anything about the likelihood of that match occurring”). See also, NRC 1992 Report, at 3-1 (“To say that two patterns match, without providing any scientifically valid estimate (or, at least an upper bound) of the frequency with which such matches might occur by chance, is meaningless”). “The party offering the evidence ‘has the burden of showing the general acceptance by experts in the field of reliability’ of that evidence. In making the determination whether the test is generally accepted, courts may properly consider not only the evidence in the record but also the reasoning and conclusions of other courts and the writings of experts.” Curnin, supra at 223 (citations omitted).
The Commonwealth did not meet this burden in the present case. The Commonwealth presented evidence that the defendant’s DNA “matched” the DNA recovered on the child’s autopsy swabbings, based on the FBI’s RFLP laboratory analysis. However, the evidence which suggested that the chance of finding another individual with the same DNA profile was 1 in 159,000 was based on the ceiling method statistical calculation which the Commonwealth failed to establish as generally accepted within the relevant scientific community. In fact, the evidence presented at the voir dire hearing and the written materials submitted on the DNA statistical calculations lead to the conclusion that the proposed ceiling method and interim ceiling method are currently the focus of great controversy within the scientific community involved.
In Commonwealth v. Lanigan, 413 Mass. 154, 162 (1992), which was decided only three months after the NRC proposed the use of the ceiling method, the Supreme Judicial Court found that there was a “lively, and still very current dispute” among scientists concerned with the role of population substructures in estimating frequencies of DNA “matches.” Lanigan held that product rule statistical calculation was not admissible under the Frye-Curnin standard. Referring to the recent 1992 NRC Report, the Supreme Judicial Court characterized the newly proposed ceiling principle as a statistical calculation which “automatically would provide for the . . . most conservative estimate of the frequency of a DNA profile” and, “is entirely in keeping with the hope that we expressed in Commonwealth v. Curnin, that the scientific community would ‘generally agree on a means at arriving at a conservative estimate of the probability of another person [173]*173having the same alleles and thus resolve all uncertainties and variables in favor of the defense.’ ” Id. at 163 (emphasis added).
However, the evidence presented in this case clearly establishes that the 1992 NRC Report did not in fact reflect a consensus of scientific opinion. Since Lanigan was decided, the relevant scientific community has not generally agreed on the ceiling principle. In addition, contrary to the hope expressed in Lanigan, some scientists involved in the national debate have found that the ceiling principle does not “automatically” provide for more conservative frequency calculations. These scientists have found that under certain conditions, the ceiling method statistical calculation can lead to less conservative frequency estimates.
Some scientists deny that genetic subpopulations exist, and feel that a ceiling principle is not necessary. Others take the position that, even if subpopulations do exist, they do not affect the statistical calculations used to interpret DNA “matches.” There are scientists who feel that a ceiling principle is appropriate, but that the databases used lead to unreliable statistical results. And some scientists feel that the ceiling principle is an artificial policy-driven compromise which is not grounded in sound science at all.
This court need not second-guess experts or decide which group of scientists proposes a more reliable statistical approach. It is abundantly clear that the relevant scientific community is engaged in a tumultuous debate over the proper statistical calculation to employ in reliably estimating the frequency of DNA profiles. “The point is not whether there are more supporters than detractors, or whether . . . the supporters are right and the detractors are wrong. The point is that there is disagreement between the . . . groups, each significant in both number and expertise (a ‘[s]ubstantial controversy,’ in the words of the NRC report). Our inquiry must end with the perception of such a disagreement and consequent lack of general acceptance.” U.S. v. Porter, 618 A.2d 629, 638 (D.C.App. Dec. 1992) (quoting People v. Barney, 10 Cal.Rptr.2d 731 (Cal.App 1 Dist. 1992)).
After reconsidering this court’s decision to allow the introduction of DNA evidence at trial, I now find that because there was no general acceptance within the relevant scientific community of the ceiling method or interim ceiling method as a reliable statistical calculation, as a matter of law, neither the evidence of the laboratory DNA “match,” nor the statistical evidence interpreting that “match,” should have been admitted at trial under the Frye-Curnin standard.
New Trial
Upon reconsideration, this court now allows the defendant’s post-conviction Rule 25(b)(2) motion for a required finding of not guilty, not as an exercise of discretion, but as a matter of law. This court chooses not to exercise any discretion which I may have at this time under this Rule. “In deciding a rule 25(b)(2) motion for a required finding of not guilty following a guilty verdict,... the judge does not properly exercise discretion concerning the weight or integrity of the evidence, but instead must assess the legal sufficiency of the evidence by the standard set out in Commonwealth v. Lattimore, 378 Mass. 671, 677 (1979). The question is one of law." Commonwealth v. Doucette, 408 Mass. 454, 456 (1990) (citations omitted).13
When a Rule 25(b)(2) motion14 is allowed as a matter of law after the discharge of the jury, a judge has three options for a remedy: 1) set aside the verdict and order a new trial, 2) order the entry of a finding of not guilty, or 3) order the entry of a finding of guilty of a lesser included offense. Doucette, supra at 455-56; Commonwealth v. Keough, 385 Mass. 314, 317-18 (1982).
Accordingly, after excluding from consideration the DNA evidence which should not have been admitted under the Frye-Cumin standard, this court viewed the evidence presented in this case in the light most favorable to the Commonwealth, and also took into account rational inferences which could be drawn from this evidence. This court now finds that, as a matter of law, the Commonwealth’s case was not sufficient to meet its burden beyond a reasonable doubt, and that the case should not have been submitted to the jury. Therefore, pursuant to Mass.R.Crim.P. 25(b)(2), this court orders a new trial. See Commonwealth v. DeBenedetto, 414 Mass. 37, 45 (1992) (retrial is remedy where an insufficiency of properly admitted evidence appears, after improperly admitted evidence is excluded from consideration).
“Evidence of this nature, based on scientific principles that every human has unique genetic characteristics and having an aura of infallibility, must have a strong impact on a jury. The erroneous admission of such evidence would undoubtedly be prejudicial in any case where, as here, the identification of the person who committed the crime is in serious dispute.” Curnin, supra at 219.
ORDER
For the reasons set forth above, the defendant’s motion after the discharge of the jury pursuant to Mass.R.Crim.P. 25(b)(2) is ALLOWED.
The guilty verdicts returned on indictment ’ No. 32440 and No. 32464 are hereby SET ASIDE.
It is further ORDERED that the defendant be granted a NEW TRIAL.