Difference between revisions of "JModelTest"
Moskalenko (talk | contribs) m (Text replace - "hpc.ufl.edu" to "rc.ufl.edu") |
Moskalenko (talk | contribs) m (Text replacement - "#uppercase" to "uc") |
||
| Line 32: | Line 32: | ||
--> | --> | ||
==System Variables== | ==System Variables== | ||
| − | * HPC_{{ | + | * HPC_{{uc:{{#var:app}}}}_DIR |
<!--Configuration--> | <!--Configuration--> | ||
{{#if: {{#var: conf}}|==Configuration== | {{#if: {{#var: conf}}|==Configuration== | ||
Revision as of 21:22, 6 December 2019
Description
jModelTest is a tool to carry out statistical selection of best-fit models of nucleotide substitution. It implements five different model selection strategies: hierarchical and dynamical likelihood ratio tests (hLRT and dLRT), Akaike and Bayesian information criteria (AIC and BIC), and a decision theory method (DT). It also provides estimates of model selection uncertainty, parameter importances and model-averaged parameter estimates, including model-averaged tree topologies. jModelTest 2 includes High Performance Computing (HPC) capabilities and additional features like new strategies for tree optimization, model-averaged phylogenetic trees (both topology and branch lenght), heuristic filtering and automatic logging of user activity.
Required Modules
Serial
- jmodeltest
System Variables
- HPC_JMODELTEST_DIR
Additional Information
To run jModelTest from the command-line load the jmodeltest module and use the 'jmodeltest' command in place of 'java -jar jModelTest.jar' command specified in the manual. You can use the jar file directly if you want, of course. It's located at $HPC_JMODELTEST_DIR/jModelTest.jar. Make sure to use the '-tr X' option where 'X' is the number of threads to use. The default is 8. Make sure your job script requests appropriate number of computing cores with the '#PBS -l nodes=1:ppn=X'.
To run a short interactive jModelTest session use the gui.rc.ufl.edu machine.