BEAST

Bayesian Evolutionary Analysis Sampling Trees or, simply put, BEAST, is an application designed for analyzing molecular sequences.

A more scientific description for its purpose would be that it provides the means to reconstruct phylogenies and also works as platform for evolutionary hypothesis testing with no conditioning on a single tree topology.

It is based on Java, a dependency that is not automatically downloaded during the installation procedure of the program. If Java is not present on the system the product does not launch.

When starting, BEAST requires some initial settings, which include providing a sample file, defining the number of chains and choosing between delta or temperature parameters.

Parsing of the can take pretty long, especially if a lower-specced computer is used. During the operation the system resources are not spared as they are put under noticeable stress, both CPU power and RAM usage.

Because of the XML format accepted by the application there is no attempt to determine the type of analysis researchers want to perform but offers the possibility to select it.

According to documentation there is a drawback to this approach as construct models may not perform well under the Markov chain Monte Carlo (MCMC) inference framework.

Also on the downside, data has to be inspected carefully because after parsing it BEAST does not come up with a straightforward answer. As such, in some cases some tinkering is necessary in order to make sure that the information has been processed correctly.

BEAST is designed as a scientific tool and it does not address the average user. Working with it is not difficult but interpreting the results and making sure that the data has been processed correctly requires solid background in molecular sequence analysis.