Multiple Alignment and Phylogenetic Tree Construction

In this exercise (except the first part) you are going to use some data files that should be in this afternoon's working directory. Change to this directory:

cd malignment

You will find that the directory is empty! The files you need are in the directory "/usr/cbs/phdcourse/malignment". You can copy all of them to your directory with the command

cp /usr/cbs/phdcourse/malignment/* .

The "*" means 'all files'. The "." at the end of the line (don't forget it!) means 'current directory'.

Now check that you have the files (there should be seven of them) with the ls command.

A tutorial about Phylogenetic Trees

Visit this site written by David E. Joyce of Clark University, Massachusetts, USA:

Phylogeny and reconstructing phylogenetic trees.

It is a well written short introduction to phylogenetic trees in general and to the distance matrix approach of reconstructing them. It includes a series of interactive programs that simulates a distance matrix from a given phylogeny and then tries to reconstruct the phylogeny from the matrix.

Note: The three solutions described on page 3 are only a subset of the distance matrix methods. They all imply a molecular clock and therefore produce rooted trees. This is not the case of the "neighbour joining" algorithm we are going to use in the next part.

Technical note: This sites uses java, a WWW programming language. We sometimes experience problems running it from some of our machines. If you do not see the phylogenetic tree displayed in the center of the intro page, run Netscape from the machine genome instead. Do the following:

Exit from Netscape.
Open a shell window where you give the commandrlogin genome.
Start a new Netscape with the command netscape &.

Technical note: There is a small bug in the java program, which means that parameter changes are ignored if you just click at the "New Tree" button. To change a parameter: click in the field you want to change, type in the new number, and press ENTER. This goes for all the subsequent pages also.

The phylogeny of HIV

You are going to:

Make a multiple alignment of the HIV gp120 protein with the program CLUSTALX.
Use this alignment to construct a distance matrix-based unrooted phylogenetic tree, also using CLUSTALX.
Construct a rooted phylogenetic tree using a program from the PHYLIP package (but based on the same multiple alignment).
Display the phylogenetic trees with programs from the PHYLIP package.

Since we have only one afternoon, we have not included any parsimony or maximum likelihood methods in this exercise.