Unlocking the Secrets of Molecular Evolution with Modeltest 3.7

Modeltest is a software tool for selecting the best-fitting model of molecular evolution
from a set of candidate models. This is done by comparing the likelihoods of different
models, which are calculated using the maximum likelihood method. The model with the
highest likelihood is considered the best-fitting model for the dataset. It was first
introduced in 2002 by David Posada and has since undergone several updates, with the
latest version being Modeltest 3.7.
Modeltest 3.7 is an update to the previous version, Modeltest 3.6, and includes several
new features and improvements. One of the main new features is the ability to perform
model selection on multiple data partitions. This allows users to run Modeltest on
different regions of a dataset, such as coding and non-coding regions, and compare the
results. Another new feature in Modeltest 3.7 is the ability to test models that include
among-site rate variation.This allows users to investigate whether there is variation in
the rates of evolution among different sites in a dataset. This is important because many
molecular datasets show evidence of such variation, and ignoring it can lead to incorrect
inferences.
Modeltest 3.7 also includes several improvements to the calculation of likelihoods,
which makes the software more efficient and accurate. Additionally, the software now
supports a wider range of models, including models for protein-coding data, and models
that allow for unequal base frequencies. Modeltest 3.7 is available as a command-line
program for Windows, Mac, and Linux operating systems and it is also available as a
GUI-based program, which allows for a more user-friendly interface.
In summary, Modeltest 3.7 is a software tool that uses the maximum likelihood method
to select the best-fitting model of molecular evolution from a set of candidate models.
The software compares the likelihoods of different models for a given multiple sequence
alignment (MSA) and the model with the highest likelihood is considered the best-fitting
model for the dataset. It uses a variety of models and allows for multiple data partitions.
Once the best-fitting model is selected, the software provides information about the
model parameters and their estimates which can be used to make inferences about the
evolutionary history of the sequences in the MSA.

References:
1: Posada D and Crandall KA 1998. Modeltest: testing the model of DNA substitution.
Bioinformatics 14 (9): 817-818.
2: Posada D and Buckley TR. 2004. Model selection and model averaging in
phylogenetics: advantages of the AIC and Bayesian approaches over likelihood ratio
tests. Systematic Biology 53: 793-
808.