Object-oriented Bayesian networks for evaluating DIP–STR profiling results from unbalanced DNA mixtures

https://doi.org/10.1016/j.fsigen.2013.09.001Get rights and content

Abstract

The genetic characterization of unbalanced mixed stains remains an important area where improvement is imperative. In fact, with current methods for DNA analysis (Polymerase Chain Reaction with the SGM Plus™ multiplex kit), it is generally not possible to obtain a conventional autosomal DNA profile of the minor contributor if the ratio between the two contributors in a mixture is smaller than 1:10. This is a consequence of the fact that the major contributor's profile ‘masks’ that of the minor contributor. Besides known remedies to this problem, such as Y-STR analysis, a new compound genetic marker that consists of a Deletion/Insertion Polymorphism (DIP), linked to a Short Tandem Repeat (STR) polymorphism, has recently been developed and proposed elsewhere in literature [1]. The present paper reports on the derivation of an approach for the probabilistic evaluation of DIP–STR profiling results obtained from unbalanced DNA mixtures. The procedure is based on object-oriented Bayesian networks (OOBNs) and uses the likelihood ratio as an expression of the probative value. OOBNs are retained in this paper because they allow one to provide a clear description of the genotypic configuration observed for the mixed stain as well as for the various potential contributors (e.g., victim and suspect). These models also allow one to depict the assumed relevance relationships and perform the necessary probabilistic computations.

Introduction

The common way to analyze DNA mixtures for forensic purposes is to use the Polymerase Chain Reaction (PCR) and STR markers [2]. This has proven to be a very successful technique, both for its speed and its high discriminating power. However, besides its many advantages, this technique has also some drawbacks. When dealing with mixtures of two contributors, for example, the method will generally not work successfully if the proportion between the DNA of the two contributors is more extreme than 1:10 [3].1 These situations are quite common, such as in cases of sexual assaults when the victim's DNA is largely predominant, or in case of microchimerism during pregnancy or following organ transplant. To address this constraint, an alternative analytical method has recently been developed and proposed [1]. This method is based on the use of a new compound marker, formed by an STR marker coupled to a marker in which a Deletion/Insertion Polymorphism (DIP) is known to be present.

DIPs as such have previously been discussed in biological and biostatistical literature (e.g., identification and characterization of di-allelic polymorphisms and allelic frequencies in particular ethnies and in natural population [5], [6], [7]), genetics (e.g., identification of DIPs as causes of genetic diseases [e.g., 8]), and forensic science (e.g., use of DIPs for analyzing highly degraded DNA [e.g., 9]). The novelty of the paper here is to present an interpretative model that represents an essential element for rendering the potential of a new compound marker formed by a DIP marker coupled to an STR marker operationally useful for practitioners. The discussion will mainly concentrate on the coherent combination of the advantages of the two kinds of polymorphism, and on how this may be formally achieved through an interpretative model. In particular, this paper aims to develop and describe a probabilistic framework for the assessment of profiling results obtained with this novel typing technique, applied in the particular context of unbalanced DNA mixtures of two contributors. The approach relies on probabilistic graphical models, in particular object-oriented Bayesian networks (OOBNs). The paper also includes a discussion of this framework for two casework examples.

Section 2 provides a short description of the DIP–STR method from a biological point of view, while Section 3 describes the generic structure of the probabilistic model (i.e., OOBN) that has been built to evaluate DIP–STR profiling results. More detailed descriptions of the different structures composing the proposed OOBN are confined to Appendix A. Section 4 presents two casework examples to illustrate the kind of calculations that can be performed with the proposed graphical network environment (i.e., to obtain likelihood ratios for particular DIP–STR profiling results). They also exemplify the flexibility of graphical models, which are readily adapted to different scenarios. The last section presents a discussion and conclusions.

Section snippets

Genetic background

The standard method for the analysis of DNA mixtures relies on STR primers as part of a procedure that seeks to amplify only selected portions of DNA, that is regions where particular STR markers are located. STR primers are only locus-specific, not allele-specific. This means that, as the DNA of both contributors have the same loci, these primers should, in theory, anneal to both the markers of the major and to those of the minor contributor. This is, in fact, what happens whenever the minor

Evaluation of DNA profiling results using graphical models

Given a mixed DNA stain from two contributors, of which only one can be taken as known (say, the victim), and a suspect who shares alleles with the stain profile in some appropriate way, two main hypotheses may generally be of interest if the evaluation is addressed at source level [13]. One, usually that referred to as the prosecution hypothesis (Hp), asserts that the mixture originates from the victim and the suspect (if the case is such that a suspect is available for comparative

General case description and DIP–STR analyses

Suppose a case in which the body of a dead women is found [1]. Circumstantial evidence leads to three suspects: a man and his two sons. Other information supports the possibility of a single perpetrator, and this information is used as an assumption in the subsequent evaluation of analytical results. A relevant blood stain – denoted A here – was collected on the victim's body. Blood of the victim and of the three suspects was also available for analysis. Using the standard protocols (autosomal

Discussion and conclusions

Unbalanced DNA mixtures are problematic for traditional STR profiling analyses, in particular when the proportion between the DNA of the two contributors is more extreme than 1:10 [3]. Cases of sexual assaults (where the victim's DNA is predominant and that of the aggressor is present only as a minor quantity) or cases of micro chimerism during pregnancy (where minute quantities of fetal DNA are present in maternal blood) are typical examples for situations in which stains of this type may be

Acknowledgements

This research was supported by the Swiss National Science Foundation, through grant no. 105311- 1445570

References (29)

  • J. Butler

    Advanced Topics in Forensic DNA Typing: Methodology

    (2011)
  • T. Clayton et al.

    Mixtures

  • Applied Biosystems

    AmpFlSTR Profiler Plus PCR Amplification Kit User's Manual

    (2012)
  • U. Vali et al.

    Insertion–deletion polymorphisms (indels) as genetic markers in natural populations

    BMC Genet.

    (2008)
  • Cited by (18)

    • Forensic molecular biomarkers for mixture analysis

      2019, Forensic Science International: Genetics
    • A solution for the rare type match problem when using the DIP-STR marker system

      2018, Forensic Science International: Genetics
      Citation Excerpt :

      The paper is structured as follows. Section 2 discusses the use of the DIP-STR marker system for extremely unbalanced mixtures, while Section 3 describes the object-oriented Bayesian network that was built to evaluate DIP-STR profiling results in Cereda et al. [7]. The chosen notation and the definition of what a full Bayesian approach to likelihood ratio assessment is, can be found in Sections 4 and 5, respectively.

    • Sensitive DIP-STR markers for the analysis of unbalanced mixtures from “touch” DNA samples

      2017, Forensic Science International: Genetics
      Citation Excerpt :

      A novel set of 10 short-sized markers has been recently developed by us to increase the number of DIP-STRs available for the analysis of forensic DNA specimens [10]. To evaluate the statistical value of investigative evidence produced by this novel type of marker, a Bayesian network model was specifically developed [11], and a theoretical comparison of the performance of DIP-STRs and autosomal STRs/Y-STRs was performed [12]. Originally developed to characterize the minor contributor in unbalanced DNA mixtures from sexual/physical offense-derived DNA samples, these markers should also be suitable to investigate other types of biological material such as “touch” DNA samples.

    View all citing articles on Scopus
    View full text