Showing papers by "Christopher M. Triggs published in 2004"

Forensic DNA Evidence Interpretation

Abstract: Biological Basis for DNA Evidence, Peter Gill and John Buckleton Historical and Background Biology Understanding PCR Profiles A Framework for Interpreting Evidence, John Buckleton The Frequentist Approach The Logical Approach The Full Bayesian Approach A Possible Solution A Comparison of the Different Approaches Population Genetic Models, John Buckleton Product Rule Simulation Testing Discussion of the Product Rule and the Subpopulation Model A Complex Case Example - DNA Evidence and Orethral James Simpson Relatedness, John Buckleton and Christopher Triggs Conditional Probabilities Joint Probabilities The Unifying Formula The Effect of Linkage Validating Databases, John Buckleton Which Is the Relevant Population? Population Databases Validating the Population Genetic Model Estimating Q Descriptive Statistics for Databases Sampling Effects, John Buckleton and James Curran Bounds and a Level Methods for Assessing Sampling Uncertainty Minimum Allele Probabilities Discussion of the Appropriateness of Sampling Uncertainty Estimates Mixtures, Tim Clayton and John Buckleton Frequentist Approaches Bayesian Approaches Statistical Evaluation of Mixtures Low Copy Number, John Buckleton and Peter Gill Changes in LCN Profile Morphology The Interpretation of LCN Profiles Non-autosomal Forensic Markers, Simon Walsh, SallyAnn Harbison, and John Buckleton Forensic Mitochondrial DNA Typing Forensic Y Chromosome Analysis Forensic X Chromosome Analysis A Famous Case Example - The Romanovs Parentage Testing, John Buckleton, Tim Clayton, and Chris Triggs Evaluation Of Evidence Paternity Trios: Mother, Child and Alleged Father Non-autosomal DNA Use of the Sub-Population Model of Balding and Nichols to Evaluate the Paternity Index Relatedness in Paternity Cases Multiple Children Inconsistencies in the Mendelian Pattern 'Exclusions' Paternity Trios: Mother, Child and Alleged Father Considering the Possibility of Silent (Null) Alleles Disaster Victim Identification, Identification of Missing Persons, and Immigration Cases, John Buckleton, Chris Triggs, and Tim Clayton Mitochondrial or Nuclear DNA? Human Remains - Obtaining a Profile from Bodily Remains Extraction of DNA from Bone, Tooth, Hair and Nail Complicating Factors DNA Intelligence Databases, Simon Walsh and John Buckleton A Brief History Functional Aspects Legislation Aspects of Forensic Significance Social and ethical considerations Interpretation Issues Associated with DNA Databases

Comment on: Why the effect of prior odds should accompany the likelihood ratio when reporting DNA evidence

Abstract: The question of how best to present evidence to a court is an open one. Clearly the area of presenting probabilistic evidence is one where the difficulties reach a high point. The matter is as old as evidence itself and is not peculiar to DNA evidence. Meester and Sjerps have reviewed previously published analyses of two very specific situations presenting evidence from DNA profiles and draw a conclusion that they then extrapolate to all evidence. This is a very bold extrapolation. We do see merits in the presentation of a posterior probability as suggested by Meester and Sjerps but these merits come with certain practical problems. These problems are recognized by Meester and Sjerps and the most beneficial outcome of this paper may be a renewal of scholarly effort in this field. We wish to assist them in this endeavour by presenting our own view on the presentation of evidence. Our view is not novel and is drawn from material that was all in the public domain before Meester and Sjerps’s publication. There is nothing inherently different about DNA evidence that sets it qualitatively aside from all other forensic evidence or even all evidence. Hence it is important that evidence derived from DNA is considered as just one form of evidence and not as something completely separate. We come immediately to the issue of setting evidence into a framework that is appropriate for court. This has been the subject of entire books by well informed authors (Robertson and Vignaux, 1995) but it is by no means settled. The issue revolves around a basic conflict: the tools best fitted to interpret evidence coherently are also those that appear to be most problematic to explain to a jury or lay person. Does the system used by the scientist have to also be the one presented in court? This is a question that is only just beginning to be asked, let alone answered. It could be argued that the presentation of scientific evidence should bend to conform to the courts’ requirements. Indeed a court can almost compel this. There have been several rulings on this subject by courts that have been used to argue for or against particular approaches to the presentation of evidence. An instance of this could be the Doheney and Adams ruling (Court of Appeal, 1996). More specifically the Doheney and Adams ruling has been, we believe erroneously, read as arguing against a Bayesian approach and for a frequentist approach. An impartial appraisal of the various methods offered for interpretation should proceed from a starting point of discussing its underlying logic. Only subsequently should it be considered how this logic may be presented in court or whether the court or jury have the tools to deal with this type of evidence. There is little advantage to the situation ‘wrong but understood’.