Comments on Assignments #4:

The book by Page and Holmes has arrived in the COOP. Also: note that textbooks are available with rebates online (Link on main page).

Clear and very readable explanations on what the different algorithms do are found in the textbooks and in the in the PHYLIP documentation. This documentation is available at Joe Felsenstein’s PHYLIP home page as part of the PHYLIP package, and as a web version here.

Ad 2) Do you get the same result with clustalx?

If you used clustalw on your unix account, the answer is no. The two versions use slightly different schemes to weight conservative substitutions.

Ad3) 3) Do you know what the reason is for the long region in the yeast ATPases that has no counterpart in the other vacuolar/archaeal ATPase catalytic subunits?

This is the first intein that was ever discovered.

Ad 4) In either clustalw or clustalx, what happens, when you change the gap penalty parameters away from their default values?

In most instances the alignment either remains unchanged or it gets worse.

Can you find a parameter combination, so that the inserted sequence present in the two yeast sequences is no longer recognizable in the alignment.

There are two directions of parameter change that will obscure the intein: A)very low gap opening costs result in a few residues sprinkled into the gap that corresponds to the intein. B) very large gap penalties force a gap placed at the beginning of yeast sequences, and the intein is aligned to the amino terminal part of the other A-subunits. The reason for this is that as a default gaps at the beginning and end of the sequences do not receive a penalty.

Ad5)

The best tree is obtained excluding positions with gaps and correcting for multiple substitutions. KEEP IN MIND THAT THE TREES SHOULD BE CONSIDERED AS UNROOTED! The two yeasts (Candida and Saccharomyces) should be together. Current conventional wisdom has the animals and Fungi forming a clade, and the plants forming a deeper branch. If one adds more V-ATPase A-subunits, one also obtains this result. This might serve as a good reminder that sometimes even branches with high bootstrap values do not hold up to closer scrutiny.

I wonder how to save a file (like the testseq1.txt you give us on

internet) under UNIX system to run Clustalw 1.7 when I use PC.

The easiest way is to FTP (Fetch on Macs) files back and forth from your PC to the UNIX machine. If you do something on a UNIX account on a regular basis, it is a good idea to keep a telnet connection running and to define a shortcut to your account in the FTP program you use.

After I get the results of the alignments and tree guide, I just know to

copy the result to creat a new *.txt under directory of Treeview, so

that I can view the tree. Is it possible to give the command to creat

the output file (*.dnd) directly to Treeview before running the

Clustalw.

Again I am not sure I understand the question; however, I can offer you a couple of comments:

A.   The dnd file created by clustalw is the tree used by clustal to calculate the multiple sequence alignment. This is not the file to get from bootstrapping (extension .phb) or neighborjoining (.ph).

B.   If you set up treeview on your pc, you can install it so that it becomes the default program to open *.dnd, *.ph and *.phb files. You can double click on these files, and they will be opened in treeview.

Could you please tell me how to download Clustlax? I just find

Clustlax on ebi site, but it didn't mention how to download it and I just

can use it on its web site.

You open a browser, and follow the link to the ebi server given in lecture 4. You click on software repository, which opens a folder with many subfolders. Select the operating system you want to run clustalx under. (If you have a PC select DOS.) Then click on the clustalw or clustalx folder (if available). In most instances the clustalx folder is a subdirectory in the clustalw folder. Open the clustalx folder (in case of the PC version you should be here and download the ZIP archive with the highest version number. Unzip the archieve and follow the instructions in the readme file (or just click on set-up).