Model comparison for PMEN1 data fit.

The table summarises the goodness of fit of four models used to the distribution of recombinations on the PMEN1 phylogeny. The maximum likelihood estimates of the parameters are given for each of the four models (see Methods and Text S1). The goodness of fit of these models is compared by the use of Akaike's Information Criterion, adjusted for finite degrees of freedom (AICc). The number of parameters in the model is denoted , and the number of degrees of freedom in the data is . The mixture model (Model 3) provides the best fit (lowest AICc) of the recombination process in the genealogical reconstruction of the PMEN1 lineage. NM = Null model (model 1); NMOD = Null model with over-dispersion (model 2); MM = Mixture model (model 3); UMM = Uncorrelated mixture model (model 4).