Taking peak area information into account when analysing STR DNA mixtures is acknowledged to be a difficult task. There have been a number of non-probabilistic approaches proposed in the literature, and some have been incorporated into computer systems, but comparatively little has been published from a probabilistic perspective. Here we briefly review our previous work on using Bayesian networks to analyse two-person mixtures within a probabilistic framework, and present preliminary results obtained for analysing two-person and three-person mixtures that combine peak area information from multiple independent samples.