OPINION
McGREGOR, Judge.I.
A jury found Kevin James Clark guilty of five class 2, dangerous offenses: attempted first-degree murder, kidnapping, first-degree burglary, and two counts of sexual assault. The charges stemmed from a brutal attack on a woman inside her Tempe apartment during the early morning of August 3, 1990. The trial judge imposed maximum sentences of 21 years on the attempted murder, kidnapping, and sexual assault convictions and ordered all sentences to be served consecutively. He imposed a presumptive sentence of 10.5 years on the first-degree burglary charge, to be served concurrently with the other sentences.
Clark filed a timely notice of appeal. He later filed a Rule 32 petition for post-convietion relief, which the trial court summarily dismissed. We consolidated Clark’s petition for review from that dismissal with his direct appeal.
Clark raises a single issue on appeal: Did the trial court err in determining that deoxy-ribonucleic acid (DNA) identification evidence was admissible under the analysis of *43Frye v. United States, 293 F. 1013 (D.C.Cir. 1923)?
II.
Prior to trial, the State forwarded items of evidence and samples of Clark’s blood and of the victim’s blood to Cellmark Diagnostic Laboratories, Inc. (Cellmark) for DNA analysis. Cellmark determined that the victim’s DNA matched DNA found on a glove and a scarf seized from Clark’s apartment. Cell-mark also concluded that Clark’s DNA matched DNA found on the glove. At Clark’s trial, a molecular geneticist testified regarding the probability of these matches occurring randomly. The witness quantified the chance that the DNA on the scarf was that of someone other than the victim as one in 87 million; the chance that the DNA on the glove was that of someone other than the victim as one in 6.9 million; and the chance that the other DNA found on the glove was that of someone other than Clark as one in 10 million.
While Clark’s appeal was pending, the Arizona Supreme Court decided State v. Bible, 175 Ariz. 549, 858 P.2d 1152 (1993), cert. denied sub nom., Bible v. Arizona, — U.S. —, 114 S.Ct. 1578, 128 L.Ed.2d 221 (1994). In Bible, the court held that the principles and theory underlying DNA testing and the criteria employed by Cellmark in declaring a match meet Frye standards. Id. 175 Ariz. at 577, 582, 858 P.2d at 1180, 1185. The supreme court also held, however, that Cellmark’s random match probability calculations are not generally accepted in the scientific community and therefore are inadmissible under Frye. Id. at 585-86, 858 P.2d at 1188-89.
Under the holding of Bible, the trial judge erred by permitting the State to introduce Cellmark’s calculations regarding the probability of a random match. The remaining inquiry for this court, as it was for the supreme court in Bible, is to determine whether the error was harmless. Id. at 588, 858 P.2d at 1191.
In Bible, the supreme court defined the standard for deciding whether admitting evidence of the probability of a random match of DNA is harmless:
Error, be it constitutional or otherwise, is harmless if we can say, beyond a reasonable doubt, that the error did not contribute to or affect the verdict. [State v.] Lundstrom, 161 Ariz. [141] 150 & n. 11, 776 P.2d [1067] 1076 & n. 11 [(1989)]. “The inquiry ... is not whether, in a trial that occurred without the error, a guilty verdict would surely have been rendered, but whether the guilty verdict actually rendered in this trial was surely unattributable to the error.” Sullivan v. Louisiana, 508 U.S. —, —, 113 S.Ct. 2078, 2081, 124 L.Ed.2d 182 (1993); accord [State v.] McVay, 127 Ariz. [450] 453, 622 P.2d [9] 14 [ (1980) ]. We must be confident beyond a reasonable doubt that the error had no influence on the jury’s judgment.
Id. at 588, 858 P.2d at 1191.
In Bible, the supreme court concluded that admission of Cellmark’s probability calculations constituted harmless error. Applying the test defined in Bible, we conclude that two material differences between this case and Bible prevent us from finding the error harmless.
The first material distinction between this trial and that involving Bible results from the difference in information the jury received about probability calculations. In considering whether the error that occurred in Bible’s trial was harmless, the supreme court stressed that experts at trial hotly debated the validity of those calculations. Id. at 588, 858 P.2d at 1191. In Bible, the trial court did not determine the admissibility of the DNA evidence in a pre-trial hearing, but rather on the basis of trial testimony. Id. at 580-81, 858 P.2d at 1183-84. As a result, the jury in Bible heard a well-qualified defense expert explain in detail the perceived flaws in the process that led to Cellmark’s probability calculations.
In contrast, the jury in this case heard no such testimony. The defense called no expert witness, so the jury heard no challenge to the validity of the probability calculations. Indeed, defense counsel barely challenged the scientific basis underlying the probability calculations during cross-examination, which *44focused instead on perceived flaws in Cell-mark’s testing of the evidence. Therefore, we have no basis for concluding, as the supreme court did in Bible, “that the picture presented to the jury did not have the aura of infallibility surrounding an unchallenged scientific theory.” Id. at 588, 858 P.2d at 1191.
The second major distinction between the two cases is that we cannot characterize the evidence of Clark’s guilt as “far beyond overwhelming,” as was true of the evidence against Bible. Id. at 589, 858 P.2d at 1192. As the dissent asserts, “there is sufficient independent and properly admitted evidence in the record to support the guilty verdiet[.]” We agree that, if the test were whether sufficient evidence exists to support the verdict, we should affirm. In Bible, however, the supreme court explained that error in admitting DNA probability match evidence is harmless only if we can “say, beyond a reasonable doubt, that the error did not contribute to or affect the verdict.” Id. at 588, 858 P.2d at 1191. The court emphasized that “[i]f the evidence against [Bible] had been closely balanced, strong, or even very strong, we think it would be impossible to say beyond a reasonable doubt that the inadmissible DNA evidence did not affect the verdict.” Id. at 588, 858 P.2d at 1191 (emphasis added). The court then went on to characterize the properly admitted evidence against Bible as “far beyond overwhelming evidence of guilt.” Id. at 589, 858 P.2d at 1192. We agree with the dissent that the evidence against Clark, set out in the dissent, is very strong. We do not think, however, that it is so overwhelming or one-sided that the jury’s guilty verdict was surely unattributable to hearing the powerful evidence of probability calculations.
The evidence provided some support for Clark’s defense. During the investigation, the victim told detectives that she thought that the assailant was her neighbor. She testified that she had observed a man, who watched the victim and her daughter for “a minute or two,” standing outside a nearby apartment watching as she and her daughter moved in several days before the assault. Clark admitted that he saw the victim when she moved in. Clark’s roommate, John Alb-ers, however, also testified at trial that he had seen the victim when she was unpacking some things from her vehicle around the time when she moved in. He testified that he greeted her, but that she did not acknowledge him. The victim was unable to identify the assailant from a pre-trial lineup that included both Clark and Albers, although at trial she identified Clark as the probable assailant.
During an initial search of the apartment Clark and Albers had shared, the police found some items of evidence, but did not recover the clothes the assailant wore during the attack. Albers testified that on the Monday or Tuesday after the assault, while Clark was in California, Albers moved many items from the apartment into Albers’ new residence. A few days later, while Clark was still in California, the police executed a search warrant at the apartment and found a plastic bag in the middle of Clark’s closet containing the clothes and shoes the assailant had worn during the attack.
The defense attempted to use this evidence linking Albers to the crime to convince the jury that the State had failed to prove Clark guilty beyond a reasonable doubt. Defense counsel also sought to defuse the DNA identification testimony by arguing that the testing process had been manipulated to obtain a positive identification of Clark. In rebuttal, the State emphasized the reliability of Cell-mark’s analysis and reminded the jury of the probability calculations provided at trial. Furthermore, testimony regarding DNA testing consumed approximately two of the seven days of trial testimony.1 Thus, the State squarely focused the jury’s attention on the DNA evidence, including the improperly admitted portion. We cannot conclude, as *45could the Bible court, that the DNA evidence “was far from the most telling part of the State’s case.” Id. at 588, 858 P.2d at 1191.
Certainly sufficient admissible evidence exists to support the guilty verdict, but other evidence tended to exculpate Clark. Admitting evidence of DNA probability calculations, however, resulted in a high potential for unfair prejudice. Under these circumstances, and given the analysis and holding of Bible, we cannot conclude beyond a reasonable doubt that the erroneously admitted DNA testimony had no influence on the jury’s verdict.
III.
We reverse Clark’s convictions and sentences and remand this matter to the superi- or court for a new trial. Because this resolution renders Clark’s Rule 32 petition moot, we deny review of the petition for review.
CLABORNE, P.J., concurs.. We do not here assert, as the dissent implies, that the DNA statistical evidence consumed two days of trial testimony. We merely note that, of the seven days of trial, approximately two days were devoted to explaining the DNA testing process, the meaning of the results, and the procedures for deriving the statistical probability calculations. In this context, we think it likely that the jury considered the DNA evidence, including the statistical probability calculations, to be very important to the issue of guilt.