Phylogenetic analysis as expert evidence in HIV transmission prosecutions

Phylogenetic analysis as expert evidence in HIV transmission prosecutions

HIV Australia | Vol. 14 No. 1 | March 2016

By Paul Kidd

Modern criminal prosecutions rely increasingly on the use of evidence from forensic science. Such evidence can help establish the identity of offenders, place defendants at the crime scene, or support a prosecution narrative.

In cases involving the alleged criminal transmission of HIV, a key challenge for the prosecution is proving a causal link between the HIV infection of the complainant and that of the accused.

No forensic test exists that, by itself, can establish such a causal link beyond reasonable doubt. Prosecutors in some cases have nonetheless been able to introduce into evidence phylogenetic analyses that purport to show a non-definitive, but inferential, link between the infections.

This article examines the use of HIV phylogenetic analysis in three Australian criminal trials. It argues that courts in Australia appear to accept forensic evidence uncritically.

As the forensic methodology used in phylogenetic analysis is inherently limited, it argues there is risk of miscarriage of justice where this type of evidence forms a substantial part of the prosecution case.

HIV phylogenetic analysis

Phylogenetic analysis is a methodology within the field of molecular virology that compares partial DNA or RNA sequences extracted from different sources to infer evolutionary relationships between them. 1 2 3 4

By identifying shared genetic sequences, an evolutionary ‘gene tree’ is constructed based on a hypothesis about the relatedness of the samples via a common ancestor. 5 6 7 8

Phylogenetics is a well-established scientific discipline, but one that is more commonly used to study viral dynamics within populations of organisms, rather than to suggest direct virological links between specific individuals.9

In criminal cases where HIV transmission is alleged, phylogenetic analysis of proviral DNA extracted from human blood is sometimes admitted in court as evidence of causation (i.e. ‘person A infected person B’), despite significant concerns about its reliability and validity.

As with DNA profiling, phylogenetic analysis can definitively rule out a connection between cases, but cannot prove a connection beyond reasonable doubt.10 11

Like DNA profiling, phylogenetic analysis compares selected gene sequences from the accused (A), the complainant (B), and a number of unrelated controls, to determine a probability that the two samples are related.

In no circumstances can the process show the direction of infection, so this is always inferred (A infected B). Nor can it rule out other routes of infection, such as that a third person (C) infected both A and B, that A infected C who infected B, and so on.12

Unlike DNA profiling, which typically uses databases with hundreds of thousands or millions of control samples to minimise error, phylogenetic studies typically use only a handful of control samples.13

The selection of these controls then becomes highly significant, as the inclusion of even one or two inappropriate controls may seriously distort the reliability of the analysis.14

A key criticism of forensic phylogenetics is that the process proceeds from a hypothesis of guilt. The analysts, who are typically not forensic specialists but research virologists or geneticists, take as their starting point a presumed relationship between the accused and the victim’s samples, and a presumed direction of infection.15

This introduces a degree of bias from which it has been argued the evidence cannot recover.16

Phylogenetic evidence is subject to the so-called ‘CSI effect’, the phenomenon whereby juries give undue weight to forensic science evidence.17

When combined with the sensation and scandal that typically accompanies HIV-related criminal prosecutions, the willingness of juries to accept such evidence is likely to be amplified by the desire to punish behaviour that carries an extreme stain of moral obloquy.

Viral forensics in Australian HIV cases

In Australia, a small number of cases involving alleged HIV transmission have employed phylogenetic analysis evidence. Three of these are described briefly below.

In F’s Case18, a 51-year-old Victorian man was convicted of three counts of conduct endangering life19 for having unprotected intercourse with three people with a disability, two of whom had tested HIV-positive.

He was sentenced to eight years’ jail, but tragically committed suicide in prison the day after being sentenced.220 At his trial, experts testified that phylogenetic analyses linked his HIV infection with that of the complainants.

According to a newspaper report, ‘DNA tests confirmed the strain to be the same as that carried by [F].’21

The technical process used in F’s Case was subsequently detailed by the investigators involved.22

They employed forensic methods originally developed for analysis of a US dentally-acquired HIV cluster23 to compare three samples taken from the accused with samples collected from four subjects whose sero-positive status was presumed to be linked (three of whom had sex with F, and the other a partner of one of the first three), and 15 ‘randomly selected’ controls.

A critical reading of their article highlights a number of issues. Only 15 control samples were used. One of the controls was a strong match for F, and was later discovered to have frequented the same sex-on-premises venue as him, raising questions about the degree of randomisation in the selection process.

While the resulting phylogenetic tree does suggest a relationship between F and the victims, it also showed a weaker, but still strong, relationship with the controls.24

In Rowland25, a 30-year-old Western Australian man was sentenced to 10 years’ jail after he was found guilty of causing grievous bodily harm to a 12-year-old boy he allegedly infected with HIV26.

The prosecution case relied in part on phylogenetic analysis, with a University of Western Australia immunologist testifying that DNA tests linked the HIV infection of the two parties.27

In Richards28, a 34-year-old Brisbane man was acquitted of charges related to the infection of an 18-year-old man. This was despite a geneticist testifying there was a ‘100 per cent probability’ that Richards had infected the man.29

Phylogenetic analysis was by no means the only evidence of causation led by the prosecution in any of these cases, and it is not suggested that any miscarriage of justice occurred.

None of the three cases went on appeal, and in the absence of published reports, it is difficult to determine what weight was given to the expert evidence or on what basis it was admitted.

But the fact that such evidence was admitted at all demonstrates a willingness by the courts to consider evidence that has limited probative value (because it cannot definitively establish causation) and has never been subject to independent scientific validation.

Admissibility of phylogenetic analyses

Expert evidence is admissible in Australian courts under an exception to the ‘opinion rule’ that permits an opinion to be given by a person with ‘specialised knowledge’ where the opinion is based wholly or substantially on that knowledge.30 31

The law conspicuously does not impose a requirement that evidence be reliable, or that the methodology used be subject to scientific validation or acceptance. Phylogenetic analysis as used in criminal trials has never been subjected to any form of independent validation or error rate quantification.

The established view is that the role of the judge in a criminal trial is to adjudicate questions of law, not fact, and that questions as to the reliability of forensic science evidence are best left to the jury.

According to this view, any concerns from the defence on such matters should be ventilated during cross-examination and via the calling of rebuttal experts.32

While this view has theoretical merit in terms of its reinforcement of the separate and complementary roles of judge and jury, in practice it creates an uneven playing field.

The state has significantly greater forensic resources to call on, and studies have found that cross examination of experts has little or no effect on juries, who tend to accept prosecution experts’ evidence without question.33

The Universal Evidence Acts do provide a mechanism for the exclusion of evidence where its probative value is outweighed by its prejudicial effect.34 Yet this safeguard has had ‘no discernible effect’ on the courts’ willingness to admit opinion evidence.35

A recent Victorian case gives some hope that the ground is shifting, but there is a long way to go.36 In the UK, after a trial involving phylogenetic analysis collapsed, prosecutorial guidelines now limit the use of the technique.37 38


HIV phylogenetic analysis is a forensic technique that has significant scientific limitations: it cannot prove a link between two samples to the required criminal standard; it provides no information about the direction of infection; it employs too few control samples; it proceeds from a hypothesis of guilt; and it has never been independently scientifically validated.

Nonetheless, evidence based on this science has been placed before magistrates and juries as ‘proof’ of facts which the scientific foundations of the technique cannot reliably support.

Problematically, Australian evidence law provides no mechanism for the exclusion of evidence based on flimsy science, preferring to leave often ill-equipped jury members to determine what weight the evidence should be given.

While the number of Australian cases employing phylogenetic evidence has to date been quite small, there is considerable cause for concern.

When combined with the moral scandal that typically surrounds HIV transmission prosecutions, phylogenetic analysis risks being given such undue weight that it could result in a serious miscarriage of justice.

Prosecutors and the judiciary should be wary of allowing such evidence to be put before juries, especially where the evidence of transmission from accused to complainant is otherwise weak; and defence counsel should work strenuously to ensure juries are made aware of the inherent limitations of the technique.


1 Andreas, B., Ouellette, B. (eds). (2001). Bioinformatics: A Practical Guide to the Analysis of Genes and Proteins. Wiley-Interscience, New York, 323ff.

2 Metzker, M., Mindell, D., Liu, X., Ptak, R., Gibbs, R., Hillis, D. (2002). Molecular Evidence of HIV-1 Transmission in a Criminal Case. Proceedings of the National Academy of Sciences of the United States of America, 99(22), 14292–14297. doi: 10.1073/pnas.222522599

3 Scaduto, D., Brown, J., Haaland, W., Zwickl, D., Hillis, D., Metzker, M. (2010). Source Identification in Two Criminal Cases Using Phylogenetic Analysis of HIV-1 DNA Sequences. Proceedings of the National Academy of Sciences of the United States of America, 107(50), 21242–21247. doi: 10.1073/pnas.1015673107

4 Machuca, R., Jørgensen, L., Theilade, P., Nielsen, C. (2001). Molecular Investigation of Transmission of Human Immunodeficiency Virus Type 1 in a Criminal Case. Clinical and diagnostic laboratory immunology, 8(5), 884–890.

5 Andreas, B. et al. (2001). op. cit.

6 Metzker, M. et al. (2002). op. cit.

7 Scaduto, D. et al. (2010). op. cit.

8 Machuca, R. et al. (2001). op. cit.

9 Bernard, E., Azad, Y., Vandamme, A., Weait, M., Geretti, A. (2007). HIV Forensics: Pitfalls and Acceptable Standards in the Use of Phylogenetic Analysis as Evidence in Criminal Investigations of HIV Transmission, HIV Medicine, 8(6), 382–387.

10 de Wit, J., Ellard, J., Murphy, D., Zablotska, I., Kippax, S. (2009). Sexual Practices, Serostatus Disclosure and Relationships: Life and Law among Gay Men’ in The Criminalisation of HIV Transmission in Australia: Legality, Morality and Reality, National Association of People Living with HIV/AIDS, Sydney, 100–107.

11 Joint United Nations Programme on HIV/AIDS (UNAIDS). (2008). International Consultation on the Criminalization of HIV Transmission: Summary of Main Issues and Conclusions. UNAIDS, Geneva, 11.

12 See diagram in Bernard et al. (2007). op. cit., 385.

13 For an example of a recent phylogenetic analysis that used just 18 control samples, see: Chen, M., Ma, Y., Yang, C., Yang, L., Chen, H., Dong, L., et al. (2015). The Combination of Phylogenetic Analysis with Epidemiological and Serological Data to Track HIV-1 Transmission in a Sexual Transmission Case, 10 PLoS ONE, 10(3), e0119989. doi:10.1371/journal.pone.0119989

14 Bernard et al. (2007). op. cit., 384.

15 Wilson, M., Weaver, S., Winegar, R. (2013). Legal, Technical, and Interpretational Considerations in the Forensic Analysis of Viruses. Journal of Forensic Sciences, 58(2), 344–357.

16 Bernard et al. (2007). op. cit., 385.

17 Wise, J. (2009). Providing the CSI Treatment: Criminal Justice Practitioners and the CSI Effect. Current Issues in Criminal Justice, 21(3), 383–399.

18 R v F (Unreported, County Court of Victoria, McInerney J, 6 Mar 1998).

19 Crimes Act 1958 (Vic) s 22.

20 Carruthers, F. (1998, 9 March). HIV Man Attempts Suicide in “deadly” Jail’. The Australian, 3.

21 Carruthers, F. (1998, 7 March). Bisexual Faces Jail for Spreading HIV. The Australian, 9.

22 Birch, C., McCaw, R., Bulach, D., Revill, P., Carter, J., Tomnay, J., et al. (2000). Molecular Analysis of Human Immunodeficiency Virus Strains Associated with a Case of Criminal Transmission of the Virus. Journal of Infectious Diseases 182(3), 941–944.

23 Blanchard, A., Ferris, S., Chamaret, S., Guétard, D., Montagnier, L. (1998). Molecular Evidence for Nosocomial Transmission of Human Immunodeficiency Virus from a Surgeon to One of His Patients. J Virol, 72(5), 4537–4540.

24 Birch et al. (2000). op. cit., 942.

25 R v Rowland (Unreported, WA District Court, Williams J, 15 Dec 1998).

26 Brook, S. (1998, 16 December). Ten Years Is Life for AIDS Sufferer. The Australian, 6.

27 Reed, D. (1998, 1 May). ‘Boy’s HIV “from Toilet Sex”’. West Australian, 3.

28 R v Richards (Unreported, Brisbane District Court, 2 May 2007).

29 Edmistone, L. (2007, 24 April). HIV Carrier Accused – “100 percent” Likely He Infected Lover’. Courier-Mail.

30 Uniform Evidence Acts [Evidence Act 1995 (Cth); Evidence Act 1995 (NSW); Evidence Act 2008 (Vic); Evidence Act 2001 (Tas); Evidence Act 2011 (ACT); Evidence (National Uniform Legislation) Act 2011 (NT)], s 79.

31 See also Dasreef Pty Ltd v Hawchar (2011) 243 CLR 588.

32 Edmond, G. (2010). Impartiality, Efficiency or Reliability? A Critical Response to Expert Evidence Law and Procedure in Australia. Australian Journal of Forensic Sciences, 83, 83–99.

33 Edmond, G., San Roque, M. (2012). The Cool Crucible: Forensic Science and the Frailty of the Criminal Trial. Current Issues in Criminal Justice, 24(1), 51–68. 51, 55–57.

34 Uniform Evidence Acts, s 137.

35 Edmond, above n 24, 88.

36 Tuite v The Queen (2015) VSCA 148 (12 June 2015).

37 R v Collins (Unreported), see: Carter, M. (2006, 9 August). Prosecution for Reckless HIV Transmission in England Ends with Not Guilty Verdict. Retrieved from:

38 Crown Prosecution Service for England and Wales, ‘Policy for Prosecuting Cases Involving the Intentional or Reckless Sexual Transmission of Infection’.

Paul Kidd is the Chair of the Victorian HIV Legal Working Group. This article is adapted in part from research undertaken in the Bachelor of Laws program at La Trobe University.