UNITED STATES DISTRICT COURT CENTRAL DISTRICT OF CALIFORNIA
June 17, 2009
UNITED STATES OF AMERICA, PLAINTIFF,
MICHAEL DENNIS WILLIAMS, ET AL., DEFENDANTS.
The opinion of the court was delivered by: Honorable Ronald S.W. Lew Senior, U.S. District Court Judge
Defendant Williams' Motion to Exclude DNA Test Results and Request for Daubert Hearing
On December 16, 2008, Defendant Williams' Motion to Exclude DNA Test Results and Request for Daubert Hearing came on for regular calendar before this Court. On December 23, 2008, this Court Denied Defendant Williams' Motion. However, this Court reserved its ruling as to whether the Government conducted LCN testing on two DNA samples found to match Defendant, and whether LCN testing meets the Daubert standard.
The Court reserved its ruling pending further briefing by the parties.
After the parties submitted additional briefing and on March 10, 2009, the Court issued an order stating that,
The critical issue in this Motion is whether the LAPD crime lab conducted Low Copy Number DNA ("LCN") testing on the two DNA samples at issue. In order to determine if LCN testing was conducted, the Court must determine the proper definition of LCN testing and if it was conducted in this case. Defendant has not made an adequate showing that LCN testing was conducted in this case; therefore, the Court requests further briefing in response to this concern.
Moreover, in Defendant's Reply he states that.03 ng and.06 ng of input DNA were tested. However, in subsequent briefings, Defendant states.3 ng and.6 ng of input DNA were tested. The Court requests that Defendant clarify his argument in this respect. (Do. No. 802.)
The Court having received and reviewed all papers submitted pertaining to this Motion, NOW FINDS AND RULES AS FOLLOWS:
The Court DENIES Defendant Williams' Motion to Exclude DNA Test Results and Request for Daubert Hearing. The Court finds that Low Copy Number DNA testing ("LCN testing") was not conducted on the two samples at issue. This determination is ascertainable from the submitted papers, and the Court does not need to hold an evidentiary hearing to make this determination. Moreover, a Daubert Hearing to determine the reliability and admissibility of LCN testing is unnecessary since LCN testing was not used in this case. ///
I. Factual Background
The LAPD recovered DNA samples #140A and #140B from latex gloves recovered at the scene of the crime. Criminologists conducted DNA testing on these samples and matched them to Defendant Williams ("Defendant"). The amount of input DNA that was amplified and available for testing for samples #140A and #140B was recorded as.6 ng and.3 ng, respectively.*fn1 This is equivalent to 600 pg and 300 pg, respectively. The manufacturer of the DNA testing kits recommends the use of between 1.0 and 2.5 ng of input template DNA in order to effectively type the DNA.
Defendant argues that testing this small amount of input DNA qualifies as LCN testing, and is therefore not admissible under Daubert. Moreover, in his Response to Prosecution's Sur-Surreply to Opposition to Motion to Exclude DNA Test Results and Request for Daubert Hearing ("Response to March 9, 2009 Order"), Defendant argues that LAPD criminologists used the slot blot method to measure the amount of input DNA, and therefore, might have tested an even smaller amount than recorded. The Government responds that the crime lab did not conduct LCN testing and the test results are reliable.
II. The Government Did Not Conduct LCN Testing
i. The Definition of LCN Testing
In order to determine if LCN testing was conducted, the Court must determine the proper definition of LCN testing.
The Government's original briefing attached the Declaration of Dr. Charlotte Word to explain LCN testing.*fn2 Dr. Word stated that LCN testing is conducted when the amount of input DNA tested is less than the testing kit's recommendation and the number of amplification cycles is thereby increased beyond the recommended number. (Govt's Sur-Surreply Word Decl. ¶5.) Dr. Word agreed that the amount of input DNA tested for the two samples was below the testing kit's recommendation. However, because the recommended number of amplification cycles was used (28 cycles), she contended that there was no LCN testing. (Govt's Sur-Surreply Word Decl. ¶6.) Moreover, Dr. Word declared that, without any increase in the amplification cycles, the amount of input DNA is low enough to constitute LCN when the amount is less than.1 ng or.2 ng of input DNA. (Govt's Sur-Surreply Word Decl. ¶7.) She stated that the two samples tested in this case were above that amount. (Id.)
In response to the Court's March 10, 2009 Order, the Government submitted another declaration submitted by Dr. Word to clarify her position. Dr. Word explains in her declaration that there are two separate ways for testing to qualify as LCN testing, which are as she previously described. (Govt's Sur-Sur-Surreply Word Decl. ¶4 and ¶5.) She expressly disagrees with Defendant's definition of LCN testing as, "...any testing for which less than the recommended amount of template DNA is being used for PCR amplification..." Moreover, she states that Defendant's expert Dr. Dan Krane's declaration that, "... The one feature that all LCN processes have in common is the knowledge or expectation that less than the recommended amount of template DNA is being used[,]" is inaccurate. (Id. at ¶6.) She agrees that any testing that qualifies as LCN testing must be below the recommended amount; however, she disputes the converse statement that all amounts of DNA below the recommended amount are LCN. (Id.)
Defendant's original briefing attached the declaration of Dr. Dan Krane to explain LCN testing.*fn3
Dr. Krane opposed the Government's definition of LCN testing and explained that there can be LCN testing without increasing the number of amplification cycles if there is a small enough amount of DNA. (Defendant's Response to Prosecution's Sur-reply to his Opposition to Motion to Exclude DNA Test Results ("Def's Response to Surreply") Decl. Krane ¶3.) Specifically, he states "Using small amounts of template DNA (even without making any changes to the testing process itself) is all that it takes for something to be in the LCN category." He supports this by citing P. Gill, et al., DNA Commission of the International Society of Forensic Genetics: Recommendations on the Interpretation of Mixtures, 160 Forensic Science International (2006) 90-101. This report states, "Low copy number is usually associated with a low amount of DNA (less than 200 pg). The method is typically associated with an elevated PCR cycle number, but it is important to realize that the effects may occur at 28 PCR cycles..."
The weight of authority supports Dr. Krane's contention that LCN testing can occur without an increase in the PCR cycle number. See also Budowle, et al., Low Copy Number -- Considerations and Caution, 12th International Symposium on Human Identification, 2001, available at http://www.promega.com/GENETICIDPROC/ ussymp12pro/contents/budowle.pdf.*fn4 Moreover, based on the supplemental briefing, it appears that Dr. Word is in agreement that LCN testing can occur without an increase in the PCR cycle number when a sufficiently low amount of input DNA is tested.
Moreover, Dr. Krane seems to agree with the Government that LCN is associated with.2 ng or less of input DNA. Dr. Krane cites authorities in his Declaration attached to Defendant's Response to Sur-reply that state that low copy number is "associated with a low amount of DNA (less than 200 pg)" and "[l]ow copy number (LCN) DNA testing typically refers to examination of less than 100pg of input DNA." (Def's Response to Surreply, Krane Decl. ¶3, ¶15.) Therefore, the Court finds that LCN testing can be defined as testing conducted on an amount of input DNA that is less than.1 ng or.2 ng. Both parties agree with this definition and the weight of scientific authority supports it. See also United States v. Davis, 602 F. Supp. 2d 658, 672 (D. Md. 2009) (finding that there would have been LCN testing if less than 100 picograms of DNA was tested).
However, Dr. Krane also states "... The one feature that all LCN processes have in common is the knowledge or expectation that less than the recommended amount of template DNA is being used." (Def's Response to Prosecution's Sur-Surreply to Opposition to Motion to Exclude DNA Test Results ("Def's Response to SurSurreply, Krane Decl. ¶6.) Dr. Krane correctly states that the two samples tested had less input DNA than the testing kit manufacturer's recommendation. The Government does not dispute this fact.
From Dr. Krane's Declaration, Defendant argues that LCN testing was used because the input DNA amount was below the manufacturer's recommendation. However, this argument overstates Dr. Krane's statements. Dr. Krane never states that failure to use the manufacturer's recommended amount makes something LCN testing. Dr. Krane merely states in the declaration attached to Defendant's Opposition Surreply that,
The one feature that all LCN processes have in common is that less than the recommended amount of template DNA is used for PCR amplification. Using small amounts of template DNA (even without making any changes to the testing process itself) is all that it takes for something to be in the LCN category. (Def's Opp. to Surreply, Krane Decl. ¶7.) This statement does not mean that this is the defining characteristic of LCN testing. In several statements, Dr. Krane states that LCN testing involves a low amount of DNA. He cites authorities that define a low amount of DNA (for LCN purposes) as less than.1 ng -.2 ng. Therefore, the Court finds that Defendant's definition of LCN testing as using input DNA below the manufacturer's recommendation is incorrect and not supported by the evidence.*fn5
Prior to the supplemental briefing the parties appeared to agree that the amount of input DNA in this case was above.1 ng -.2 ng. If the amount of input DNA was in fact.1 ng -.2 ng, then the testing conducted would not fit the above definition of LCN testing.
However, Dr. Krane continues on to give another definition of LCN testing. He quotes, Dr. Bruce Budowle, stating "LCN typing is better defined as the analysis of any results below the stochastic threshold for normal interpretation." (Def's Opp. to Surreply, Krane Decl. ¶4.) He then states, "Using less than the manufacturer's recommended amount of input DNA can result in testing results that are at or below the stochastic threshold." (emphasis added). (Id.) The problem is that he is vague in defining how much less than the amount recommended could lead to this or whether the testing in this case was at or below the stochastic threshold.*fn6 On one hand, his statement suggests an amount below the recommendation could suffice. Especially, considering Budowle's quote that, "LCN typing has been defined by Gill, et al. (4,5) as typing samples containing less than 100pg of DNA. However, because of variation in estimates of DNA quantity within and between methods of quantitation, LCN typing is better defined as the analysis of any results below the stochastic threshold for normal interpretation." (emphasis added).
On the other hand, Dr. Krane quotes Butler, Forensic DNA Typing, Second Edition: Biology, Technology and Genetics of STR Markers (Academic Press 2005) p.68) which states,
When amplifying very low levels of DNA template, a phenomenon known as stochastic fluctuation can occur. Stochastic effects, which are an unequal sampling of the two alleles present from a heterozygous individual, result when only a few DNA molecules are used to initiate PCR. PCR reactions involving DNA template levels below approximately 100 pg of DNA... have been shown to exhibit allele dropout.
This would indicate that the low amount at issue in LCN testing is still less than 0.2 ng. Finally, it is important to note again, that Dr. Krane never declares that LCN testing was used in this case.
Dr. Krane's additional definition of LCN testing as testing an amount of DNA that falls beneath the stochastic threshold is consistent with the definition that LCN testing is the testing of a low amount of input DNA (between 100-200 pg).*fn7 This definition attempts to account for any error in measuring the amount of input DNA by not defining LCN by an alleged certain amount of input DNA. When insufficient input DNA is tested (as in the case of LCN testing), the result is stochastic fluctuation and allele drop-out. Therefore, test results with a certain amount of stochastic fluctuation and allele drop-out indicate that a low amount of input DNA was used (i.e. an amount below the stochastic threshold). Therefore, Dr. Krane's definition that LCN testing is testing an amount of DNA that falls beneath the stochastic threshold accounts for the situation where the amount of input DNA was below the manufacturer's recommendation, measured and recorded as above 100-200 ng, yet actually improperly quantified.
Therefore, the weight of evidence presented does not support Defendant's over-inclusive definition of LCN testing that qualifies all tests below the manufacturer's recommendation as LCN testing. Instead, LCN testing is only defined as testing an amount of input DNA below.2 ng or testing an amount below the somewhat amorphous stochastic threshold. Therefore, the testing in this case will only qualify as LCN testing if the amount of input DNA tested was below.2 ng or shown to be below the stochastic threshold.
ii. The Quantiblot Method Properly Quantified the Input DNA
Determining the amount of input DNA tested is critical to determining whether LCN testing was conducted in this case. Defendant's supplemental briefing argues extensively that the Quantiblot method used to quantify the amount of input DNA is not reliable and the samples were not properly quantified.
Defendant argues the stated amounts of samples 140-A, 140-B and 141-A of.3 ng,.6 ng, and 1.5 ng, respectively, cannot be taken at face value. Dr. Krane explains that these amounts could be inaccurate because "... given the subjectivity and lack of sensitivity of the Quantiblot methodology used to quantitate the samples in this case, the amount of DNA for these samples cannot be reliably established." (Def's Response to Sur-Surreply, Krane Decl. ¶5). He states, "The values stated in the DNA Input column are only a rough estimate of the total amount of input DNA in a sample and do not, for mixed samples 140-A, 140-B, and 141-A, reflect the input DNA for each component of the mixture." (Id. at ¶8.) In addition, "the accuracy of the values stated in the DNA Input column are dependent upon the accuracy of the Quantiblot method of quantitation, and on the knowledge of the exact amount of the sample that is used in the PCR process." (Id.)
Thus, he argues that the input DNA value for the samples is only accurate if the concentration of DNA was properly determined and the stated 20 ul (microliters) of solution for the reaction was actually used.*fn8 (Id.) He argues that there is no evidence that the concentration was properly determined because there is no documentation that 20 ul of solution was used. Instead, it is merely assumed that amount was used because it is standard protocol to use 20 ul. (Id.)
In addition, he states that the Quantiblot methodology relies on a visual comparison that is subject to subjectivity. (Id. at ¶10.) He states that most labs currently use a more reliable method called real time PCR. (Id. at ¶9.)
Further, he explains that there are two types of detection methods used with the Quantiblot kit: Colorimetric or Chemiluminescence. (Decl. 10). Both of these methods use a visual, non-quantitative comparison. (Id.) According to the LAPD SID Standard Operating Procedures, the Colorimetric method is used for samples that have greater than.15 ng of human DNA. (Id.). Thus, he argues that the Quantiblot method of analysis was unreliable in this case because the Colorimetric method was used to quantify Samples 140-A and 140-B, which had less than.15 ng of human DNA. (Id. Para 12.)
However, the Court finds Dr. Word's declaration persuasive in demonstrating that Dr. Krane has misinterpreted the amount of DNA quantified under the Quantiblot method and is incorrect in his statement that there is no evidence that 20 ul of solution was used.
To begin, Dr. Krane's argument that there is no evidence that 20 ul of solution was actually used for the reaction is contradicted by the Amplification Record (Govt's Sur-Surreply Exhibit D). First, the LAPD SID standard operating procedures call for 20 microliters (ul) of solution. Second, examining the Amplification Record, it is apparent that 20 ul was used in the calculation because each sample has its concentration level listed along with the DNA input level. The DNA input level is determined by multiplying the DNA concentration per microliter by the amount of solution. If one examines Exhibit D, each DNA input level is the product of the DNA concentration level per microliter multiplied by 20 ul.*fn9
Most importantly, Dr. Krane's argument that Colorimetric detection is not reliable for samples 140A and 140B because they had less than 0.15 ng of human DNA is flawed. Section 8.1.5 of the LAPD SID Standard Operating Procedures states, "Colorimetric detection can be used for samples that have greater than 0.15-ng of human DNA." Dr. Word explains that this refers to the amount of input DNA (which is the right column on the Amplification Record. (Govt's Sur-Surreply Word. Decl. ¶19.) Moreover, this translates to a concentration level of 0.015 ng/ul when 10 ul of a sample is assayed (which is the middle column). (Id.) Reviewing the Amplification Record sample 140-A, 140-B and 141-A are 0.03 ng/ul, 0.015 ng/ul, and 0.15 ng/ul, respectively. Each sample is above the required 0.15 ng required by the LAPD SID Standard Operating Procedures. (Id.) In other words, "Dr Krane has mistakenly reversed the level of detection of the amount of DNA on the slot blot with the DNA concentration of the samples being tested when he 'calls into question the use of the colorimetric method of detection below the 0.15 ng level.'" (Id.) Therefore, Defendant's argument that the slot blot concentration was incorrectly quantified, and therefore unreliable fails.
Finally, Dr. Krane seems to suggest that the Quantiblot method is unreliable and has been abandoned for a supposedly more reliable method of quantitation called Real Time PCR. It is true that technology has produced a more advanced method to quantify the DNA; however, "the Quantiblot was accepted in the scientific community in 2004 when these tests were performed." United States v. Davis, 602 F. Supp. 2d 658, 669 n.15 (D. Md. 2009).*fn10
Based on this evidence, Defendant has not shown that the DNA was improperly quantified under the Quantiblot Method. As a result, there is no evidence that the input DNA for Samples 140-A, 140-B, and 141-A contained less than 200 pg of input DNA or were below the stochastic threshold. Thus, the Court finds that LCN testing was not conducted in this case and DENIES Defendant's Motion. See also Davis, 602 F. Supp. 2d at 672 (finding that even though the exact quantity of input DNA present in the samples was not known with certainty, there was no evidence to conclude that less than 100 pg of input DNA was tested, and therefore there was no LCN testing).
IT IS SO ORDERED.