Director of Public Prosecutions v Tuite (Ruling No 3)
[2017] VSC 442
•11 August 2017
| IN THE SUPREME COURT OF VICTORIA | Not Restricted |
AT MELBOURNE
CRIMINAL DIVISION
S CR 2014 0007
| THE DIRECTOR OF PUBLIC PROSECUTIONS | |
| v | |
| CLINTON TUITE | |
---
JUDGE: | HOLLINGWORTH J |
WHERE HELD: | Melbourne |
DATE OF HEARING: | 22 and 23 June 2017 |
DATE OF RULING: | 11 August 2017 |
CASE MAY BE CITED AS: | DPP v Tuite (Ruling No 3) |
MEDIUM NEUTRAL CITATION: | [2017] VSC 442 |
---
EVIDENCE – Admissibility – Expert evidence – DNA evidence – Application to exclude prosecution DNA evidence – Probative value – Whether admission of DNA evidence would give rise to unfair prejudice to the accused – Application refused – Evidence Act 2008 (Vic) s 137.
---
APPEARANCES: | Counsel | Solicitors |
| For the DPP | Dr N Rogers QC | Solicitor for Public Prosecutions |
| For the Accused | Mr J Desmond | Doogue O’Brien George |
HER HONOUR:
Background
The accused is charged with aggravated burglary, attempted rape, indecent assault, and causing injury intentionally. All of the charges arise out of an incident which occurred in the early hours of 13 October 2007, at the home of the complainant.
The primary issue at trial will be the identity of the offender. The prosecution seeks to rely on DNA evidence, derived from items located at or near the complainant’s home, to prove that the accused was the offender. DNA evidence is conventionally presented in the form of a likelihood ratio, which expresses in mathematical terms the likelihood that a particular person contributed to the detected DNA sample as compared to a person chosen at random. In calculating the likelihood ratios for the DNA samples, scientists at the Victoria Police Forensic Science Service used a statistical software program called STRmix.
STRmix was developed by scientists in Australia and New Zealand. It is used in New Zealand, and in most Australian States and Territories. It was introduced into Victoria in 2013.
Various aspects of the DNA evidence have been the subject of earlier applications and interlocutory appeals:
(a) Over 23 days between July and October 2014, Emerton J heard applications by the defence to exclude the prosecution DNA evidence, under either ss 79 or 137 of the Evidence Act 2008 (Vic). At the heart of both of the applications was the proposition that the STRmix methodology was a “new” science, that was largely untested, and had not been generally accepted by the forensic science community. On 19 December 2014, Emerton J dismissed the defence applications (in Ruling No 1).[1] On 12 June 2015, the Court of Appeal dismissed an interlocutory appeal against Ruling No 1.[2]
(b) Over three days in May and June 2015, Emerton J heard an application by the prosecution to limit the scope of the evidence that could be given by Ms Jane Taupin, a biological scientist proposed to be called by the defence. On 25 June 2015, Emerton J ruled (in Ruling No 2) that Ms Taupin did not have the specialised mathematical or statistical knowledge to enable her to express some of the opinions about STRmix to which the prosecution took objection.[3] On 16 July 2015, the Court of Appeal refused the defence leave to appeal against Ruling No 2.[4]
(c) Over three days in April 2017, Emerton J held a Basha hearing, to enable the defence to cross-examine two key prosecution DNA experts, Dr Duncan Taylor and Professor David Balding.
[1]R v Tuite [2014] VSC 662.
[2]Tuite v R [2015] VSCA 148 (“CoA No 1”).
[3]R v Tuite [2015] VSC 303.
[4]Tuite (No 2) v R [2015] VSCA 180.
On 2 May 2017, Lasry J listed the charges for trial commencing on 23 October 2017. That was the fifth time the charges had been listed for trial; previous trial dates had been vacated due to the earlier applications and interlocutory appeals.
In the current application, which was argued before me on 22 and 23 June 2017, the defence sought to exclude the evidence of Dr Taylor and Professor Balding under s 137 of the Evidence Act because:[5]
(a) Of the contents of the “PCAST materials”, to which I shall refer shortly; and/or
(b) STRmix utilises something called the “product rule”, in a particular circumstance which the defence say is inappropriate.
[5] Section 137 of the Evidence Act provides that the court in a criminal proceeding must refuse to admit evidence adduced by the prosecutor, if its probative value is outweighed by the danger of unfair prejudice to the accused. The probative value of evidence means the extent to which the evidence could rationally affect the assessment of the probability of the existence of a fact in issue.
At a directions hearing on 21 July 2017, I indicated that I would be refusing that application, for reasons to be published at a future date. These are those reasons.
Before considering those two grounds, it is necessary to briefly describe the DNA evidence and the STRmix methodology.
The prosecution evidence
The prosecution seeks to rely at trial on the results of the analysis of DNA derived from the following items, which were seized by police at the crime scene:
(a)Items 1-1, 1-2 and 1-3, being three samples taken from a makeshift blindfold, which the offender had placed over the complainant’s eyes; the blindfold was found on the floor outside the complainant’s laundry;
(b)Item 4-1, being a trace sample taken from seven cable ties, of the type which had been used to restrain the complainant’ hands; the cable ties were found on the complainant’s bathroom floor;
(c)Item 5A-1, being a portion of a filter from one cigarette butt found on the ground of a neighbouring property; and
(d)Item 5B-1, being a portion of a filter from a second cigarette butt found on the fence of that neighbouring property.
The prosecution analysis is that Items 1-1, 1-2 and 1-3 contain a mixed DNA profile from three contributors, Item 4-1 contains a partial mixed DNA profile from at least two contributors, and Items 5A-1 and 5B-1 both show a single source contributor.
Using STRmix, the results from the DNA analysis of those six samples yielded extremely high likelihood ratios, which point to the accused as being a contributor (and in the case of items 5A-1 and 5B-1, being the sole contributor) to the Items.
DNA analysis is more complex when there is more than one contributor to a sample,[6] and when the DNA is of low quality[7] or quantity.[8] Emerton J discussed the problems in a case such as this in the following terms:
If DNA on a crime scene sample is of low quality or quantity, then the profile may be ambiguous. Low-template DNA profiles are more likely to exhibit ‘stochastic’ or random effects complicating their interpretation. These affects may include ‘drop in’ which are peaks in the profile that are not true alleles and, conversely, alleles may go missing or ‘drop out’ of the profile producing partial profiles only.[9]
[6]As is the case of the samples other than Items 5A-1 and 5B-1.
[7]Item 4-1 is “low-template” DNA, in that the amount analysed is significantly less than the optimal amount of 0.5 ngs. Some of the contributors on the items with multiple contributors also produced very low amounts of DNA.
[8]Items 1-1 and 4-1 produced only partial or incomplete profiles.
[9]Ruling No 1 at [19].
Her Honour went on to explain how STRmix works, and how it purports to account for many of those random effects in a DNA profile.[10] She also explained the difference between a probabilistic, fully-continuous methodology (such as STRmix) on the one hand, and a binary methodology (which is what was used historically) on the other. It is unnecessary for present purposes to reproduce those paragraphs, or to refer to them in any detail.
[10]Ruling No 1 at [20]-[29].
The prosecution proposes to call a number of witnesses to give evidence in relation to different aspects of the DNA evidence. The two experts who are relevant for the purposes of the current application are Dr Taylor and Professor Balding. Dr Taylor is the Principal Forensic Scientist (Forensic Statistics) at Forensic Science SA, and one of the co-inventors of STRmix. Professor Balding is a Professor of Statistical Genetics at the University of Melbourne and at University College London. In general terms, they will both give evidence about the mathematical and statistical models underpinning STRmix, and the validation of probabilistic genotyping software.
The PCAST ground
The status of the PCAST materials
PCAST is the President’s Council of Advisors on Science and Technology. It is an American body, which describes itself as:
an advisory group of the Nation’s leading scientists and engineers, appointed by the President to augment the science and technology advice available to him from inside the White House and from cabinet documents and other Federal agencies. PCAST is consulted about, and often makes policy recommendations concerning, the full range of issues where understandings from the domains of science, technology, and innovation bear potentially on the policy choices before the President.[11]
[11]The quote comes from the preface to the PCAST report.
In September 2016, PCAST published a lengthy report entitled “Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods.” Most of the 160-page report was concerned with matters of general scientific methodology and validity. One of the 10 chapters, chapter 5, was headed “Evaluation of Scientific Validity for Seven Feature-Comparison Methods”. The relevant one of those seven methods was dealt with over eight pages of the report, under the heading “5.2 DNA Analysis of Complex-mixture Samples.”
On 6 January 2017, PCAST published a short document entitled “An Addendum to the PCAST Report on Forensic Science in Criminal Courts.” The addendum said that it had been developed “to address input received from stakeholders” since the publication of the PCAST report. It specifically considered feedback it had received in relation to five of the seven methods, one of which was DNA analysis of complex mixtures.
The PCAST report and addendum (collectively, “the PCAST materials”) appear to represent the collective work of a large number of scientists from various fields, as well as judges and lawyers. It is not possible to tell who authored or contributed to any particular part, or what (if any) qualifications any author(s) or contributor(s) had in relation to any particular field of science (including DNA analysis).
During the Basha hearing in April 2017, the defence put the PCAST materials to Dr Taylor and Professor Balding in cross-examination. Those witnesses agreed with some, and disagreed with other, opinions expressed in the PCAST materials. The refusal of this exclusion application will not prevent the defence from cross-examining prosecution witnesses about the PCAST materials at trial.
The PCAST materials were marked for identification for the purpose of the current voir dire application. Whether the defence will be able to call anyone at trial, to give evidence in support of the opinions expressed in the PCAST materials, remains to be seen.
When the matter first came on before me for directions on 25 May 2017, the defence had not yet made any attempt to contact anybody from PCAST, to see if they could find a suitably-qualified expert who could give relevant evidence.
For the purposes of the voir dire, I proceeded on the assumption that the PCAST materials represented the evidence that a suitably-qualified defence expert would be able to give. Although it was hypothetical to proceed on that assumption, it seemed to be the fairest and safest way to proceed: it would allow the defence to have their application (and any further interlocutory appeal) heard and determined, whilst minimising the risk that the trial date would have to be adjourned once more, in the event that the defence did manage to obtain a late expert report supporting the opinions expressed in the PCAST materials.
In fact, at a directions hearing on 21 July 2017, after the completion of the voir dire, defence counsel informed me that, after making numerous inquiries of persons who may have been involved in the preparation of the PCAST materials, he thought it unlikely the defence would be able to call any such witness (either by video-link or in person).
The contents of the PCAST materials
Chapter 5.2 of the PCAST report began with a general description of the problems inherent in analysing complex mixtures of DNA, and the importance of properly calculating the statistical weight to be given to probabilities.
The PCAST report then noted that initial approaches to the interpretation of complex mixtures relied upon subjective judgment by examiners, together with the use of simplified statistical methods such as the ”Combined Probability of Inclusion” (“CPI”). The authors were highly critical of the subjective judgment approach and the use of CPI, concluding that the method was clearly not foundationally valid.
The PCAST report then discussed various probabilistic genotyping computer programs that have been developed in order to bring some objectivity to the analysis. The report noted that as at March 2014, at least 8 such programs had been developed, including STRmix. The report described such programs as clearly representing a major improvement over purely subjective interpretation, but said they still required careful scrutiny to determine:
(a) Whether the methods are scientifically valid, including defining the limitations on their reliability; and
(b) Whether the software correctly implements the methods.
The report said that appropriate evaluation of the proposed methods should consist of studies by multiple groups, not associated with the software developers, that investigated the performance and defined the limitations of programs by testing them on a wide range of mixtures with different properties. It then identified four particular issues that it said should be addressed as part of such evaluation, before stating as follows:
A number of papers have been published that analyze known mixtures in order to address some of these issues. Two points should be noted about these studies. First, most of the studies evaluating software packages have been undertaken by the software developers themselves. While it is completely appropriate for method developers to evaluate their own methods, establishing scientific validity also requires scientific evaluation by other scientific groups that did not develop the method. Second, there have been few comparative studies across the methods to evaluate the differences among them – and, to our knowledge, no comparative studies conducted by independent groups.
Most importantly, current studies have adequately explored only a limited range of mixture types (with respect to number of contributors, ratio of minor contributors, and total amount of DNA). The two most widely used methods (STRmix and TrueAllele) appear to be reliable within a certain range, based on the available evidence and the inherent difficulty of the problem. Specifically, these methods appear to be reliable for three-person mixtures in which the minor contributor constitutes at least 20 percent of the intact DNA in the mixture and in which the DNA amount exceeds the minimum level required for the method.
For more complex mixtures (eg more contributors or lower proportion), there is relatively little published evidence….
When further studies are published, it will likely be possible to extend the range in which scientific validity has been established to include more challenging samples.[12]
[12]At pp 80-81 of the PCAST report.
At the end of Part 5.2, the PCAST report made the following relevant findings:
Finding 3: DNA analysis of complex-mixture samples
Foundational validity. PCAST finds that:
(1) [CPI based methods] …
(2) Probabilistic genotyping. Objective analysis of complex DNA mixtures with probabilistic genotyping software is relatively new and promising approach. Empirical evidence is required to establish the foundational validity of each such method within specified ranges. At present, published evidence supports the foundational validity of analysis, with some programs, of DNA mixtures of 3 individuals in which the minor contributor constitutes at least 20 percent of the intact DNA in the mixture and in which the DNA amount exceeds the minimum required level for the method. The range in which foundational validity has been established is likely to grow as adequate evidence for more complex mixtures is obtained and published.[13]
[13]At p 82 of the PCAST report.
Most of the 9-page PCAST addendum addressed matters of general principle, relating to all types of feature-comparison methods of expert evidence. PCAST re-affirmed the general principle that there is no justification for accepting that a method is valid and reliable in the absence of appropriate empirical evidence. The addendum then re-examined five of the seven methods evaluated in the PCAST report, one of which was DNA analysis of complex mixtures. The discussion of DNA analysis in the addendum was brief – only seven paragraphs.
The addendum described its previous findings on DNA analysis as follows:
PCAST found that empirical testing of [probabilistic genotyping] had largely been limited to a narrow range of parameters (number and ratio of contributors). We judged that the available literature supported the validity and reliability of [probabilistic genotyping] for samples with three contributors where the person of interest comprises at least 20% of the sample. Beyond this approximate range (ie with a larger number of contributors or where the person of interest makes a lower than 20% contribution to the sample), however there has been little empirical validation.[14]
[14]At p 8 of the PCAST addendum.
In the original report, PCAST had said that it was the minor contributor who had to contribute not less than 20% of the DNA to the sample, whereas the addendum said it was the person of interest who had to make that minimum contribution. There is nothing in the addendum, or other materials before the court, that explains why that change was made. Apart from that change, the addendum repeated that available literature supported the validity and reliability of probabilistic genotyping in certain circumstances.
The addendum briefly mentioned submissions that had been made to it by several interested parties, including the STRmix developers, before concluding:
The path forward is straightforward. The validity of specific [probabilistic genotyping] software should be validated by testing a diverse collection of samples within well-defined ranges. The DNA analysis field contains excellent scientists who are capable of defining, executing and analyzing such empirical studies.
When considering the admissibility of testimony about complex mixtures (or complex samples), judges should ascertain whether the published validation studies adequately address the nature of the sample being analysed (eg DNA quantity and quality, number of contributors, and mixture proportion for the person of interest).[15]
[15]At p 9 of the PCAST addendum.
Submissions concerning the PCAST materials
The defence submitted that the PCAST materials stood for the proposition that although probabilistic genotyping software may be what was variously described as “a good science”, a “developing science” or “a promising science”, it should not be relied upon by this court until sufficient independent or empirical studies had been carried out by persons independent of the developers of it. The defence submitted that PCAST was weighty authority for the proposition that no such studies had been done in relation to any probabilistic genotyping software, including STRmix. It was submitted that the prosecution DNA evidence should not be seen as highly probative, because “an august body” such as PCAST had put “a real cloud” over probabilistic genotyping.
There is no question that the tenor of all of the PCAST materials was that independent empirical studies should be undertaken in relation to all types of forensic science, including DNA analysis of complex samples.
However, defence counsel essentially wanted the court to ignore those parts of the PCAST materials that stated that the available literature did support the validity and reliability of probabilistic genotyping by STRmix, in certain circumstances (namely, with three contributors, where the person of interest comprises at least 20% of the sample). At the Basha hearing, Dr Taylor and Professor Balding did not accept that there was any sound reason for PCAST to have imposed that particular limitation; they also noted that no explanation had been given for making the change from “minor contributor” to “person of interest”. However, Professor Balding and Dr Taylor both clearly agreed with the broader PCAST proposition, namely, that the validity and reliability of some software, including STRmix, had been established in some circumstances. Defence counsel sought to rely on the Taylor/Balding criticisms of the limitation, to persuade me to reject or ignore all of the parts of the PCAST materials that said that the validity and reliability of any probabilistic genotyping software had been established in any circumstances. That submission involved selectively picking the parts of the PCAST materials which suited the defence case, and ignoring the parts that did not.
The prosecution also submitted that the parts of the PCAST materials which were critical of the state of validation studies for probabilistic genotyping software should be given limited weight, because the PCAST materials either pre-dated, or failed to have regard to, relevant independent validation studies. Examples given by prosecution witnesses on the Basha hearing included: a published, independent study by the FBI, and comparative exercises performed by the forensic regulator in the UK and the European network of forensic science agencies.
In essence, the defence submission was that the prosecution DNA evidence had little probative value, because the PCAST materials seriously undermined its reliability.
That submission must fail, for several reasons. First, as discussed above, the PCAST materials do accept the validity of STRmix in certain specified circumstances,[16] and are nowhere near as critical of it as the defence suggested. Secondly, even if the PCAST materials were construed in the artificial way contended for by the defence, there is prosecution evidence that PCAST may be out-of-date in its understanding of the state of the empirical research, in what is a rapidly-evolving area. Thirdly, and most importantly, in assessing the probative value of the DNA evidence, the court is now required to assume the reliability of that evidence.
[16]In the present case, Items 1-1, 1-2 and 1-3 are all three-person samples, according to the prosecution experts; the only one of those which would not meet the 20% threshold suggested by the PCAST addendum would be Item 1-2. The other samples have fewer than three contributors.
At the time of Ruling No 1 (and the appeal therefrom), when assessing the probative value of evidence, a Victorian court was only required to assume the truthfulness, but not the reliability, of the evidence.[17] Accordingly, in assessing the probative value of the prosecution DNA evidence, Emerton J considered the defence’s four grounds of attack on the reliability of that evidence.
[17]Dupas v The Queen (2012) 40 VR 182.
One of those four grounds of attack was that the evidence was alleged to be unreliable because of “[t]he lack of consensus in the broad DNA scientific community as to the reliability of STRmix, the lack of comparison with other statistical methodologies or independent review or assessment, and the paucity of validation studies, particularly given the low level amounts of DNA in issue.”[18] Emerton J accepted that those were all matters than went, potentially, to the reliability of the likelihood ratios generated by STRmix.[19] However, she noted that, like most of the reliability arguments raised by the defence, there was a contest between the expert evidence to be called by either side, which would be for the jury to resolve.[20]
[18]Ruling No 1 at [85(b)].
[19]Ruling No 1 at [91].
[20]Ruling No 1 at [92]-[98].
Emerton J carefully evaluated all of the evidence relating to those attacks on the reliability of the DNA evidence, before concluding that the DNA evidence viewed as a whole was highly probative:
It may be used by a jury to put the accused both inside and outside the house on the night in question. This is so, notwithstanding only small amounts of DNA matching that of the accused were found on the relevant items inside the house, and that other people also contributed to the DNA found on these items. The limitations in the STRmix methodology acknowledged by the prosecution witnesses must have some effect on the quality of the DNA evidence. However, I am not persuaded that they erode its probative value to any significant degree. Whilst the amounts of DNA may be small in some cases, the fact that DNA matching the accused’s was found on a number of items both inside and outside the house in my view fortifies the overall probative value of the DNA evidence, which I assess to be high.[21]
[21]Ruling No 1 at [112].
The Court of Appeal held that “that conclusion was well open on the evidence before her Honour.”[22]
[22]CoA No 1 at [123].
The attack based on the PCAST materials is, to some extent, just another iteration of the ground discussed at [40] above.[23] Even if I was not required to assume the reliability of the evidence for the purposes of this application, no good reason has been put forward which would cause me to come to a different conclusion to her Honour.
[23]The other three grounds on which reliability was attacked before Emerton J related to: the alleged risk of contamination of the samples; whether validation studies by the amplification kit manufacturer supported its use for low copy DNA; and some specific problems that Ms Taupin identified with the DNA analysis of each of the Items.
Since the appeal from Ruling No 1, the High Court determined in IMM v The Queen,[24] that in assessing the probative value of evidence for the purposes of s 137, a trial judge should assume that the jury will accept the evidence at its highest and, thus, should not have regard to the credibility or reliability of the evidence.
[24][2016] HCA 14.
Defence counsel sought to rely upon the following passage from the majority reasons in IMM:
It must also be understood that the basis upon which a trial judge proceeds, that the jury will accept the evidence taken at its highest, does not distort a finding as to the real probative value of the evidence. The circumstances surrounding the evidence may indicate that its highest level is not very high at all. The example given by JD Heydon QC was of an identification made very briefly in foggy conditions and in bad light by a witness who did not know the person identified. … [I]t is an identification, but a weak one because it is simply unconvincing.[25]
[25]IMM at [50], French CJ, Kiefel, Bell and Keane JJ.
But the circumstances surrounding the prosecution’s DNA evidence in this case is in no way analogous to Mr Heydon QC’s foggy night example.
For all of those reasons, I find that the prosecution DNA evidence has high probative value.
The next matter to be considered is whether the PCAST materials have given rise to any risk of unfair prejudice and, if so, whether that outweighs the probative value.
Evidence is not unfairly prejudicial to an accused merely because it makes it more likely that they will be convicted. Unfair prejudice may arise in a number of ways; for example, if the evidence leads the jury to adopt an illegitimate form of reasoning, or to give the evidence undue weight.
The defence’s primary complaint was that Dr Taylor and Professor Balding had, during the Basha hearing, referred to unpublished studies, which they had not produced or identified with sufficient clarity. In particular, Professor Balding disagreed with the proposition put to him in cross-examination (and taken from the PCAST report) that there had been few comparative studies conducted by independent groups. He said that he had taken part in several comparative exercises by the European network of forensic science agencies, and that the forensic regulator in the UK had also done a comparative exercise between STRmix and TrueAllele.
The following exchange then occurred between defence counsel and Professor Balding:
Can you identify those studies for me, please? --- Well, unfortunately, as I said, I am not aware they have been published, but they were just used by these agencies for internal reasons, and they were circulated to the different groups, but they were conducted independently, because the body, the Forensic Regulator in the UK conducted one of the studies.
I want my independent witness to review what these other noble persons say? --- Yes. In any case, I broadly agree that there have not been enough comparative studies, there have been some, and those that have, do exist, I’m not aware of any that have been published, which is unfortunate. I have been involved in enough that I can form, you know, my own opinion based on comparisons, but all of this information is not public, yes, which is unfortunate, yes.[26]
[26]Transcript of 19 April 2017, p 85.
I queried whether Professor Balding would have been anticipating being asked for details of that information at that time, to which defence counsel replied “Well, he probably wasn’t.”[27] Defence counsel conceded that the defence had sought no further detail of those comparative studies, either from Professor Balding or the prosecution. That is a matter that can be followed up by either side prior to trial. At trial, the experts on both sides can be asked to identify or produce information to support their respective opinions. If they are unable to do so, the other side can invite the jury to give little or no weight to that particular opinion. A failure to provide suitable supporting evidence can also be the subject of an appropriate direction or warning by the trial judge, if necessary. I am not persuaded that there has been any unfairness in relation to his aspect, that is not capable of being remedied prior to, or addressed at, the trial.
[27]Transcript of 22 June 2017, p 46.
Defence counsel tried to raise several other matters, which were said to give rise to a danger of unfair prejudice. Some were abandoned, after I pointed out that they were no more than an attempt to re-agitate matters that had already been ruled upon by Emerton J, and which had nothing to do with PCAST.
One was a slightly different point, albeit that it had nothing to do with PCAST. Defence counsel argued that the so-called “CSI effect” would give rise to the possibility of the jury placing undue weight on the prosecution DNA evidence. In any case involving scientific evidence, the trial judge must be alert to the need to ensure that the jury are able to understand and assess that evidence, and to give such warnings or directions as may be needed to ensure that the jury do not give undue weight to scientific evidence.
In Ruling No 1, Emerton J rejected a defence argument that there was a risk of unfair prejudice because the DNA evidence would be too difficult for a jury to understand. Her Honour concluded that any perceived prejudice, that might arise from the complexity of the evidence, could be addressed by a strong direction that the jury must not act on the STRmix conclusions unless they are wholly satisfied that they are soundly based on the evidence that they have heard and understood.[28] The defence have not explained why strong directions, of the type envisaged by Emerton J, would not be sufficient to counteract any “CSI effect” in this case – particularly given that this is a case in which it will be clear to the jury that there are differing scientific opinions.
[28]Ruling No 1 at [124].
For these reasons, I am not persuaded that the evidence of Dr Taylor and Professor Balding should be excluded under s 137 on the basis of the PCAST materials.
The product rule
The defence also seeks the exclusion of that same evidence under s 137, on the basis that STRmix utilises something called the “product rule” in the case of allelic drop-outs.
In his supplementary statement dated 17 April 2015, Dr Taylor set out the mathematical equations used in the STRmix program, and explained the various assumptions that underlie those equations.
STRmix applies a well-established mathematical rule called the “product rule” to probabilities; what that means is that the probabilities are multiplied. The defence do not criticise the general use of the product rule in STRmix, only its use in relation to allelic drop-outs.
The defence proposes to call evidence from Professor Ranajit Chakraborty, a statistician and geneticist from the University of North Texas Health Science Centre. At paragraph [21] of his report dated 22 July 2016, Professor Chakraborty was critical of the use of the product rule in relation to allelic drop-outs. He set out his understanding of how STRmix takes allelic drop-out into account, and criticised what he believed to be a lack of acceptance and validation for that treatment.
In a further supplementary report, filed on 2 September 2016, Dr Taylor responded to the Chakraborty report. According to Dr Taylor, Professor Chakraborty’s comments at paragraph [21] demonstrated a misunderstanding of the way in which STRmix takes allelic drop-out into account. Furthermore, even if STRmix operated in the manner asserted by Professor Chakraborty, Dr Taylor said there would be widespread acceptance and validation of such an approach.
In cross-examination during the Basha hearing, Dr Taylor and Professor Balding both explained why they disagreed with Professor Chakraborty’s criticism of the use of the product rule in relation to allelic drop-out.
It is not necessary, for present purposes, to consider in more detail the competing expert opinions in this regard. Clearly, they are matters to be explored thoroughly at trial. However, for the purposes of s 137, I am required to assume the reliability of the prosecution evidence.
I also reject the defence suggestion that the circumstances surrounding this part of the prosecution’s DNA evidence are analogous to Mr Heydon QC’s foggy night example, which I discussed at [45] above.
For the reasons discussed earlier, the DNA evidence has high probative value. The defence have not pointed to any separate risk of unfair prejudice which might flow from the admission of this part of the prosecution evidence.
For these reason, I am not persuaded that the evidence of Dr Taylor and Professor Balding should be excluded under s 137, on the basis of the use of the product rule in relation to allelic drop-out.
2
1