Koichiro Tamura, Daniel Peterson, Nicholas Peterson, Glen Stecher, Masatoshi Nei, and Sudhir Kumar (2011) MEGA5: Molecular Evolutionary Genetics Analysis using Likelihood, Distance, and Parsimony methods. Molecular Biology and Evolution.
(to be submitted)
Kumar S, Dudley J, Nei M & Tamura K (2008) MEGA: A biologist-centric software for evolutionary analysis of DNA and protein sequences. Briefings in Bioinformatics 9: 299-306.
- Download PDF
Tamura K, Dudley J, Nei M & Kumar S (2007) MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Molecular Biology and Evolution 24: 1596-1599.
- Download PDF

 

 

Tests of the Reliability of a Tree Obtained

In this example, we will conduct two different tests using protein-coding genes from the chloroplast genomes of nine different species.

Ex 4.0.1:  Start MEGA by double-clicking on the MEGA desktop icon, or by using the Windows start-menu to click on the MEGA icon located in the programs folder.

Ex 4.0.2:  Activate the data in the Chloroplast_Martin.meg file by using the File|Open command.

We will begin with the bootstrap test for the neighbor-joining tree.

Ex 4.1.1:  Select the Phylogeny |Bootstrap Test of Phylogeny|Neighbor-Joining Tree command from the main application menu.

Ex 4.1.2:  An analysis preferences dialog box appears. Use the Models pull-down to ensure that the Amino Acid|p-distance model is selected. Note that only the Amino Acid submenu is available.

Ex 4.1.3: Click “Compute” to accept the default values for the rest of the options. A progress indicator provides the progress of the test as well as the details of your analysis preferences.

Ex 4.1.4: Once the computation is complete, the Tree Explorer appears and displays two tree tabs. The first tab is the original Neighbor-Joining tree, and the second is the Bootstrap consensus tree.

Ex 4.1.5: To produce a condensed tree, use the Compute|Condensed Tree menu command from the Tree Explorer menu. This tree shows all the branches that are supported at the default cutoff value of BCL >= 50.

Ex 4.1.6: To change this value, select the View|Options menu command and click the cutoff values tab. Select the Compute|Condensed Tree menu command, and the NJ tree will reappear.

Ex 4.1.7: Print this tree. (see Ex 3.3.1 - Ex 3.3.2)

Ex 4.1.8: Select the File|Exit Tree Explorer (Ctrl-Q) command to exit the Tree Explorer. A warning box will inform you that your tree data has not been saved. Click “OK” to close Tree Explorer without saving the tree session.

For neighbor-joining trees, you may conduct the standard error test for every interior branch by using the Phylogeny|Neighbor-Joining command. In MEGA, this test is available for the p-distance, Poisson Correction, and Gamma distance for amino acid sequences. Since we did the above analysis for the p-distance, we will use the same distance estimation method to compare the results from the bootstrap and standard error tests.

Ex 4.2.1: Go to the Phylogeny menu and select the Construct Phylogeny|Neighbor-Joining command to produce an analysis preferences dialog box. In the Models preference pull-down, be sure that p-distance is the model chosen. Click on the Test of Phylogenytab to reveal the test options. Under the Test of Inferred Phylogeny option group, check the Interior Branch Test option.

Ex 4.2.2: Click “Compute” to begin the computation. A progress indicator will appear briefly. The neighbor-joining tree with confidence probabilities (CP) from the standard error test of branch lengths is displayed on the screen.

Ex 4.2.3: Compare the CP values on this tree with the BCL values of the tree that you printed in the previous procedure.

Ex 4.2.4: Now exit MEGA using the Alt + X command.