Statistical analysis of DNA mixtures with artefacts

by Therese Graversen

DNAmixtures is a statistical framework for analysis of DNA samples from one or multiple donors. It implements the statistical model introduced in Analysis of Forensic DNA mixtures with Artefacts (published with discussion in Journal of the Royal Statistical Society, series C).

DNAmixtures has been used to give evidence in several cases in both Denmark and the United Kingdom. Contact Therese Graversen for inquiries related to the use of DNAmixtures in criminal casework.

Screenshot of DNAmixtures in the R GUI
Common forensic applications
Computation of likelihood ratios
Deconvolution of the DNA mixture
Combined analysis of multiple traces of DNA
Visual assessment of the statistical modelling
Does the hypothesis explain the data well?
Is the distribution of peak heights adequate?
Efficient exact computation
Runs on a standard desktop or laptop
Allows a higher number of contributors
A complete statistical framework
Exact computation of the likelihood function
Maximum likelihood estimation of parameters
Simulation of genotypes and peak heights
Access to various conditional distributions of peak heights given data

System requirements

DNAmixtures may be installed on any system (Windows, Linux, or MacOS), on which the following three programs are installed:

RStatistical software
HuginBayesian network software
RHuginAPI for calling Hugin from within R

For an installation guide, please see the installation page.

Now available: Hugin-free lite-version of DNAmixtures

The DNAmixturesLite package is intended as a service to enable users to try DNAmixtures without purchasing a commercial licence for Hugin. When at all possible, we strongly recommend the use of DNAmixtures rather than its lite-version. While the lite-version seeks to provide the full functionality of DNAmixtures, note that computations are much less efficient and that there are some differences in available functionality.


Questions and comments are very welcome and can be directed to the author, Therese Graversen.