Sentence examples for model selection measures from inspiring English sources

A closer examination of the results shows that this is due to the model selection measures incorrectly selecting the +I+G ASRV where the true ASRV is +G.

Table 5 shows that when the original model amino acid frequencies were randomly perturbed, there was a definite trend among all of the model selection measures to select the corresponding '+F' version of the particular model over the original models.

In the following subsections, we outline how we investigated the effects of various properties of amino acid alignments on the three non-nested model selection measures (AIC1, AIC2, BIC) when applied to protein model selection.

When the sequence length is increased to 1000 characters, the difference between the likelihoods of the +G and +I+G models increases and is enough for the model selection measures to prefer the +I+G models.

The Akaike information criterion (AIC) [ 27] and Bayesian information criterion (BIC) [ 28] belong to a different class of model selection measures that compare all of the models simultaneously according to some measure of fitness.

While the focus of this work has been to demonstrate the usefulness of performing protein model selection, it must be stated that model selection measures can only provide information on which of the given set of models best-fits the data and cannot give any indication of how close a particular model is to reality.

The BIC is another model selection measure and is equivalent to selecting the model with the maximum posterior probability and is calculated from BIC = -2 ln L i + N i ln n,     (2) where n is the sample size (sequence length).

The AIC is a popular model selection measure that attempts to strike a balance between the goodness-of-fit and complexity of a model.

To measure the discriminative power of a particular model selection criterion, we run a ProBMoT experiment where the given criterion is used to rank the models.

Model selection criterion.

First, we established data partitions for each alignments using a greedy search in PartitionFinder v1.0.1 (Lanfear et al. 2012), with linked branch lengths, constraining the models of evolution to those available in RAxML, and using AICc for model selection (a measure of AIC corrected for small sample sizes, Hurvich and Tsai 1989).