Forensic samples are among the most complex encountered. Blood is best known, but other biological matrices also carry genetic information. Cheek swabs (buccal swabs) collect cells from the inside of the mouth and have the advantage of being a non-invasive sample collection compared to a blood draw. Hair, depending on the presence of the root, is amenable to DNA typing. Semen, vaginal fluids, and vaginal swabs are collected in sexual assault cases. Any surface on which biological fluids (blood, oral fluid, vaginal fluid, etc.) are deposited becomes a potential DNA source.

The initial deposition (called the primary transfer) occurs from a person to a surface. It is the deposition of blood, saliva, semen, or other biological substance directly from the body onto a surface. This process could be a victim’s blood dripping onto an assailant’s clothing, saliva on a cigarette, or seminal fluid on a bedsheet.

This chapter focuses on model evaluation and selection in Hierarchical Modelling of Species Communities (HMSC). It starts by noting that even if there are automated procedures for model selection, the most important step is actually done by the ecologist when deciding what kind of models will be fitted. The chapter then discusses different ways of measuring model fit based on contrasting the model predictions with the observed data, as well as the use of information criteria as a method for evaluating model fit. The chapter first discusses general methods that can be used to compare models that differ either in their predictors or in their structure, e.g. models with different sets of environmental covariates, models with and without spatial random effects, models with and without traits or phylogenetic information or models that differ in their prior distributions. The chapter then presents specific methods for variable selection, aimed at comparing models that are structurally identical but differ in the included environmental covariates: variable selection by the spike and slab prior approach, and reduced rank regression that aims at combining predictors to reduce their dimensionality.