Mixture Likelihood Ratio
Definition
The ratio of the probability of observing a mixed DNA profile if the person of interest is a contributor to the probability of observing the same profile if they are not. A Bayesian network computes this by marginalising over all unknown contributor genotypes.
- Compares
- Prosecution vs defence contributor hypotheses
- Computation method
- Bayesian network marginalising over unknown genotypes
- Output form
- A single ratio, often reported in scientific notation
- Depends on
- Chosen number-of-contributors assumption
Common questions
Why does the number of assumed contributors change the likelihood ratio?+
Adding an extra unknown contributor increases the number of genotype combinations that could explain the mixture, which typically dilutes the weight of evidence, so analysts must justify the contributor number before computing the ratio.
Can two different software tools give different likelihood ratios for the same mixture?+
Yes, because tools differ in their underlying statistical model, stutter and drop-out parameters, and validation data, so laboratories validate a specific tool and report which one produced the figure.
Related terms
- Belief Propagation
- An algorithm for computing marginal and posterior probabilities in a Bayesian network by passing messages between neighbouring nodes. Exact on tree-structured networks;...
- Conditional Probability Table (CPT)
- A table that specifies the probability distribution of a node given every combination of states of its parent nodes. Every non-root node...
- D-Separation
- A graphical criterion that determines whether two sets of nodes in a Bayesian network are conditionally independent given a third set. If...
- Directed Acyclic Graph (DAG)
- A graph in which edges have a direction (from parent to child node) and no path can return to a node it...
- Sensitivity Analysis
- A technique for assessing how much the posterior probabilities in a Bayesian network change when the values in the conditional probability tables...