Understanding complex DNA mixtures
We are exploring the strengths and weaknesses of algorithms and software programs used to interpret complex DNA mixtures (samples that contain DNA from more than one person).
What we are doing
There are multiple analysis programs designed to make sense of complex DNA mixtures encountered as part of criminal investigations. We are attempting to find out the strengths and weaknesses of the different programs.
Why we are doing it
It is vitally important that we have confidence in the analysis programs that are being used to try and identify whose DNA may be present in a complex mixed sample of DNA recovered and analysed during a criminal investigation. No one has undertaken a global assessment of all of the available analysis programs before using the same dataset. This allows us to compare the analysis programs with each other directly and report accurately on the limitations and error rates of the current methods.
How we will do it
We have designed a set of 820 DNA mixtures from known donors at known concentrations and known numbers of contributors. We will run all the samples against all the software packages available and will perform statistical analyses on the resulting data. The aim being to report on the effectiveness of the programs to separate out the correct contributors from the mixture.