Opinion
March 29, 1995
J. Bruce Maffeo, New York City, for Kevin Ladson, defendant.
Robert M. Morgenthau, District Attorney of New York County, New York City (Peter M. Kougasian of counsel), for plaintiff.
The indictment against Kevin Ladson and Antoine Atterbury charges each defendant with six counts of murder in the second degree in connection with the strangulation deaths of Nikki Silas and Dalia Kabira "Brittany" Rojas, whose bodies were discovered at about 3:15 P.M. on December 19, 1992 in apartment 5A at 211 West 109th Street in Manhattan. Counts 1 and 2 allege that the defendants intentionally caused the deaths of Ms. Silas and Ms. Rojas; counts 3 and 4 are felony murder charges, causing the deaths of Ms. Silas and Ms. Rojas during the commission of the crime of rape in the first degree; and finally, counts 5 and 6 also involve felony murder, causing the deaths of Ms. Silas and Ms. Rojas during the commission of a robbery.
I
PROCEDURAL BACKGROUND
After a lengthy pretrial hearing, defendants' Huntley and Mapp motions were denied; however, each defendant in separate extensive statements implicated the other in various ways. The prejudice resulting therefrom could not be eliminated by appropriate redactions, and, thus, pursuant to Bruton v United States ( 391 U.S. 123), the case was severed for purposes of trial.
The People seek to proceed first against Kevin Ladson. The prosecution's evidence consists primarily of Mr. Ladson's own statements, as well as witnesses and evidence that place both defendants at the scene of the murders. Since the only surviving "eye-witness" to the actual killings is the noncooperating codefendant, the case against Mr. Ladson is largely circumstantial although there is other potential evidence in the case. The Medical Examiner took swabs from Nikki Silas' vaginal area. These samples contained genetic material from Ms. Silas herself and from other unknown sources. The Medical Examiner also took scrapings from under Ms. Silas' fingernails.
The People seek to offer at trial DNA test results generated by using the restriction fragment length polymorphism (RFLP) technique. The FBI laboratory that performed the analysis concluded to a high degree of probability that the DNA typing pattern for certain of the evidence specimens recovered from the vaginal area of Ms. Silas' body matched the DNA typing pattern of the blood of defendant Kevin Ladson provided pursuant to an order of this court.
II
DEFENDANT'S MOTION REGARDING THE DNA ANALYSIS
(A) RFLP TESTS
The defense, while not directly challenging the "match" between the male DNA found in Ms. Silas' vaginal area and Mr. Ladson's own DNA pattern, nevertheless seeks to preclude the introduction of the results of the People's RFLP analysis on the following grounds: that the "match" between Ms. Silas' own blood and her own vaginal fluids exceeds generally accepted margins of variation, and that the failure of this standard control technique renders the entire analysis (including the otherwise acceptable match between Mr. Ladson's blood and the male sample found in Ms. Silas' vaginal area) inconclusive.
III
RFLP ANALYSIS
(A) SCIENTIFIC BACKGROUND
Most sections of the DNA molecule are similar in all individuals; however, certain of the noncoding sites are "polymorphic" meaning that they vary from individual to individual. The areas of DNA where there is a great deal of variability in the repeated sequences of the base pairs of nucleotides are called variable number tandem repeats (VNTRs).
The substitution of even one of the four nucleotides in the sequence of DNA is a variation that can be detected by "restriction enzymes" used to cut DNA. The restriction or cutting enzyme will cut a sample of DNA into fragments whose length depends on the location of the cutting sites it recognizes as containing a specific sequence of letters. The length of a DNA section that contains the repeated sequences is a direct function of the number of times the letter sequence is repeated before it is cut. Length differences in DNA that can be detected and measured following digestion or cutting are termed restriction fragment length polymorphisms.
Thus DNA testing using the forensic RFLP technique does not measure an individual's genetic code but measures and compares the length of hypervariable restriction fragments, VNTRs. Because VNTRs vary greatly in length from individual to individual, the assumption is that there is a low likelihood that any two people will "match". To reduce the likelihood of a coincidental match, the three or four different VNTR regions "probed" are on different chromosomes.
An x-ray or audiograph reflects the distribution of DNA fragments by length after they have migrated in a gel under the stimulation of an electric current. The position of the dark bands on the autorad reflects the length of the VNTRs, the fragments that vary from individual to individual. The bands on the audiograph are studied and compared to determine if there is a "match". Where the suspect's (known) band has a position different than the recovered evidence (unknown) band, he is excluded as a contributor. Where the suspect's band is determined by visual and computer analysis to be similar to the recovered evidence band within predetermined limits, there is said to be a "match", i.e., the suspect may be the source of the recovered evidence.
(B) THE FBI'S RFLP PROCEDURES
The defense does not challenge the basic laboratory techniques employed by the FBI. However, the defense argued at the hearing that the lack of blind proficiency testing raised questions about the over-all integrity of the FBI test results. The court however has concluded that the issue of error rates and quality control goes to the weight, not the admissibility of the test results.
(C) THE EXCLUSION OF CODEFENDANT ANTOINE ATTERBURY
It is the People's contention that absent any indication to the contrary, the only possible sources of the semen found on Ms. Silas are Mr. Ladson and Mr. Atterbury. As far as Mr. Atterbury is concerned, both the People's expert, Dr. Deadman, and the defense expert, Dr. D'Eustachio, agree that an examination of the autorads reveals that Mr. Atterbury (K2) could not be the source of the male fraction of the dried secretions (Q2) and vaginal swab (Q4).
IV
THE DEFENSE ARGUMENT THAT "BAND SHIFTING" IN MS. SILAS' KNOWN SAMPLES CALLS INTO QUESTION THE ENTIRE FBI ANALYSIS
The FBI has concluded that the K3 known Ladson sample matches the male fraction of the unknown samples recovered from Ms. Silas' vaginal area, Q2 and Q4. Dr. Deadman testified that before the FBI declares a match, the evaluator must make both a visual assessment of the bands and a quantitative measurement, i.e., "the difference between the two bands must not be greater than five percent of the mean or average of the two bands" (transcript, at 564). Furthermore, each locus that is examined using a different probe would have to match under both the visual and quantitative criteria.
The defense does not dispute that the match between the known Ladson sample (K3) and the unknown male fraction of samples Q2 and Q4 is well within the FBI's (and generally accepted) standards. The defense contends, however, that because the DNA fragments derived from Ms. Silas' own blood (K1) exceed the generally accepted margins of variation when compared with the female fractions of the evidence specimens (Q1, Q2 and Q4) recovered from swabbing performed on Ms. Silas, the entire FBI analysis is of dubious reliability and should not be admitted. In essence the defense argues that an essential control has failed; that is, known sources which would be expected to match have failed to match. The defense further argues that these discrepancies suggest the existence of technical flaws in the administration of the tests and that consequently it is difficult if not impossible to draw any possible known/unknown matches.
The People argue that these discrepancies which are highlighted by the defense are not significant variations, and in any event, there is a simple scientific explanation for this perceived "band shifting". The known blood sample (K1) taken from Ms. Silas was extracted during an autopsy performed the day after her death and the People contend it has been scientifically established that blood tends to degrade after death. The K1 band has characteristics that denote degradation but do not reflect technical unreliability. Vaginal fluids suffer less postmortem degradation and thus no similar band shifting is observed as to the vaginal samples. Finally, the People argue, the variations between Ms. Silas' blood and her vaginal fluids remain constant regardless of which genetic locus is observed, clearly demonstrating that the observed variation is a consequence of degradation of the blood sample and not the result of any irregularity in the FBI's testing procedures.
The defense expert, Dr. D'Eustachio, testified that the tests performed by the FBI in this case revealed that a disproportionate number of comparisons between blood taken from the body of Ms. Silas and the female fraction of evidence retrieved from her body fell either outside or at the outer limit of the range or "window" set by the FBI itself. Dr. D'Eustachio pointed to data in the FBI's own study (Budowle, Fixed Bin Analysis for Statistical Evaluation of Continuous Distributions of Allelic Data from VNTR Loci, for Use in Forensic Comparisons, Am J Hum Genet 48:841 [1991] [hereinafter the Fixed Bin paper]) that led to the establishment of its match window.
Dr. D'Eustachio testified that comparison of samples known to share a common origin operates as an important internal control of the reliability of a test and he indicated his belief that the People's own witness at the hearing, Dr. Deadman, had espoused the same position in sworn testimony in other cases. Finally he stated that his conclusion that the failure of such a control renders the test results inconclusive is consistent with the prevailing view in the scientific community at large and the position taken in the NRC Report that "when a control indicates irregularities in an experiment, the result in question must be considered inconclusive" (National Research Council, DNA Technology in Forensic Science, at 55 [hereinafter NRC Report]).
The People initially addressed the issues of controls in this particular experiment. Dr. Deadman pointed out that there was good correspondence between the Q2 male fraction DNA, the Q4 male fraction DNA and the K3 known sample DNA from defendant Ladson. Dr. Deadman indicated that the similarity in the two male fractions in this case is an indication of the reproducibility of the procedures in this instance. Dr. Deadman further pointed out that the FBI (and other laboratories) use an internal control in all instances. The laboratory uses the best quality DNA to serve as a control at all phrases — electrophoresis, Southern Blotting, identifying what probe is being used — because the control DNA must give a predictable pattern. If the control profile in a test result is not in the designated position and the bands do not fall within a particular size range, the results are not used. The FBI uses the K562 cell line as a control. In order for a particular probe to be used at all, the K562 band in the control lane must show up in predetermined positions. Since the control K562 DNA is good quality DNA, there should be little or no band shifting and Dr. Deadman testified that the results were all within the expected range.
Because of these internally consistent results, the People contend that the defense argument of a failure of internal controls should be rejected. More importantly, the People contend, the band shifting observed in the Silas blood sample is a scientifically explainable phenomenon.
The People's expert, Dr. Deadman, testified that there are always some differences in the results of DNA tests from the same known source. In this case, Dr. Deadman testified, there were two types of band shifting, degradation shifts and a concentration shift. The concentration shift results in a distorted shape of the band which is, Dr. Deadman testified, consistent with the large amount of DNA in the known sample DNA from Ms. Silas' blood (K1). The extensive shifting in the female known sample (K1) is caused by a combination of degradation and concentration causing the DNA in that lane to run faster than the female fraction DNA, which is not degraded and which does not contain as much DNA.
There are objective indicators of degradation in the DNA. Dr. Deadman testified that a "yield gel" or test gel will indicate degradation because a degraded sample will contain smaller DNA samples. Thus even before autorads are developed, degradation is evident in the yield gel from streaking at or below where the discrete band would appear. Dr. Deadman pointed out at the hearing the high degree of streaking in the K1 (blood) lane and very little streaking in the Q2 lane developed from dried secretions on Ms. Silas' body. Similarly, the Q4 lane, developed from a vaginal swab, showed no streaking. "The more streaking you have, the more degradation [is] involved" (transcript, at 572). The degradation noted in the K1 sample is consistent with the literature. Dr. Deadman pointed out that the "Bar Paper" (Bar, Postmortem Stability of DNA, Forensic Science Intl 39 [1988], at 59-70) documents the existence of postmortem degradation of DNA in blood samples; "in some cases, DNA degradation was already prominent after a short period" (id., at 59).
Dr. D'Eustachio, the defense expert, acknowledged the "Bar Paper's" discussion concerning degradation of blood from cadavers; however, Dr. D'Eustachio felt that the paper was not helpful in predicting whether or to what degree degradation will appear in a corpse's blood. Dr. D'Eustachio further conceded that the Fixed Bin paper (Budowle, op. cit.), a well-known article in peer review literature, established that bands in lanes of a gel that contain degraded DNA will tend to migrate farther than bands in a lane without degraded DNA and that this phenomenon was familiar to forensic scientists.
While Dr. D'Eustachio conceded that degradation was possible in the Silas blood sample and that degradation can result in band shifting, the defense nevertheless argues that in one instance the shift exceeded the range (or "match window") in the Fixed Bin study (and in other instances approached the high range of the Fixed Bin study) and thus the entire FBI analysis should be declared inconclusive.
The People urge that, given the internal controls in the FBI's procedures and the demonstrable predictability that band shifting will occur where degradation exists, the defense argument should affect the weight of the evidence but not its admissability — a classic question for the jury to decide. Both sides look to the discussion of band shifting in the NRC Report (op. cit., at 60-61) to support their respective positions; the People point to the statement that "[b]and shifting could cause two DNA samples from one person to show different patterns" (id., at 60) and the defense notes the Report's recommendation that labs "adopt the policy of declaring samples that show apparent band shifting to be `inconclusive'" (id., at 61).
The court is satisfied that the issue here is one of weight, not admissibility. Each side's expert has noted that the band shifting in the control lanes (known to known) was so extensive that it exceeded the relevant match window. The People have demonstrated that they followed established procedures and that the other controls here reveal no irregularities. They are entitled to an opportunity to convince the jury that this phenomenon is the predictable and scientifically explainable result of degradation of the DNA extracted from a corpse and not a result of flawed testing techniques.
[Portions of opinion omitted for purposes of publication.]