Integrative Biology 200A University of California, Berkeley PRINCIPLES OF PHYLOGENETICS Spring 2008 Lab 5: Alignment In this lab we’re going to try to look at the effects of different methods of DNA alignment. We’ll try different settings in ClustalW. We’ll go through how to convert the output from ClustalW to a Nexus file, and how to manipulate that Nexus file. We’ll also use POY, which does alignment and tree searching together. Trying different alignment strategies and seeing how they affect your phylogenetic hypothesis is a great way to add some interest to your project, since it helps explore the question of just how much changes at this step affect the trees you build. Get the sequences You have three options: 1) If you want you can get sequences from the organism that you will be working on for your project. Go to http://www.ncbi.nlm.nih.gov/ search for your organism in the nucleotide data base, and try to find about fifteen sequences of a gene from different organisms that will be useful for your study. Download them to your desktop in FASTA format. 2)If you already know that there is nothing good on genbank (or you’re just in a rush), There’s an example FASTA file of COI sequences from some Cephalopod species that you can download from http://ib.berkeley.edu/courses/ib200a/cephalopod_COI.txt. 3) For the really lazy you can just download the Conus sequence file from http://ib.berkeley.edu/courses/ib200a/sequences.fasta. I won’t hold it against you. ClustalW ClustalW is the most commonly used program for multiple sequence alignments. It works by first doing pairwise alignments of each pair of sequences to calculate distances between them. It uses those distances to derive a tree. That tree then guides the production of the multiple sequence alignment. For this lab, we’re going to access ClustalW directly through web page of the European Bioinformatics Institute, where you can control each of the search parameters. Since ClustalW is the industry standard, we’re going to use it for comparison to POY, and to see the effects that parameters can have on the alignment. -Go to http://www.ebi.ac.uk/Tools/clustalw2/index.html. For the first pass let’s just use the defaults. -From the output format menu select ‘aln wo/numbers’. Hit the Browse button and select your FASTA file. Then hit run.A web page will come up, letting you know that your job is in progress. It will automatically refresh, so that when your job is finished, a new web page will come up. -Pay attention to how long it takes, and when the job is finished, make a note of the alignment score. There will be four files available for you to download. The output file is a description of the alignment process. The .aln file is your actual sequence alignment in Clustal format. The .dnd file is the guide tree used to make the alignment. The input file is the file that you uploaded to ClustalW. -Download the alignment and guide tree files and give them unique names. -Repeat the same alignment only this time run a ‘fast’ rather than a ‘full’ alignment. Don’t forget to select ‘aln wo/numbers’. It probably won’t make a difference for this alignment, because you don’t have a very big matrix, but for more species or a longer sequence a fast alignment can lead to a big savings in time. The ‘fast’ alignment works by only considering a chunk of the sequence at a time. The options on the second line refer only to the ‘fast’ alignment and have to do with how big a chunk of the sequence you consider. How long did it take? Did it seam faster to you? Is your alignment score as good as it was last time? -Save the alignment file again, but don’t bother with the guide tree. Let’s do this one more time, but this time let’s give it really unrealistic parameters to see how the parameters do effect the alignment. -Set the gap open penalty to 1 and the gap extension penalty to 5. This will make it easier to start a gap then to extend it, a highly unrealistic situation. -Run the same sequences and save the alignment file. Don’t forget to select ‘aln wo/numbers’. Comparing the Alignments You can use Mesquite to open up the alignments you made with different input parameters and see how much the results changed. You may also wish to export the resulting alignments for use with either Phylip or Nona/TNT and see whether the different alignments produce changes in the phylogenetic reconstruction of these clades. POY POY is a program by Ward Wheeler, David Gladstein, and Jan De Laet that is freely available on the web (http://research.amnh.org/scicomp/projects/poy.php). It views the alignment as a problem, in which gaps and deletions are events that happen along the tree. In other words, POY implements a way to build the alignment and the tree at the same time. As its creators describe it: “An essential feature of POY4 is that it implementsthe concept of dynamic homology allowing optimization of unaligned sequences…. POY4 provides a unified approach to co-optimizing different types of data, such as morphological and molecular sequence data. In addition, POY4 can analyze entire chromosomes and genomes and take into account large-scale genomic events (translocations, inversions, and duplications).” As a result, the output from POY is a tree, not an alignment per se. -Create a folder on your desktop called ‘POY folder’. Drag the unaligned sequence file into this folder. -Find poy4.exe (Start>Programs>poy4>poy4.exe) and mol1.txt (Start>Programs>poy4>docs>mol1.txt). Copy and paste both of these into your new folder. The docs folder also contains a file called “tutorial commands” or something like that. You can open this folder if you want and copy and paste all of the commands below. -Open mol1.txt using a text editor. You’ll see that it’s just a series of input sequences, not aligned yet. Close it. -Open poy4 by double-clicking on it. An interface will appear that has three separate boxes. The box on the lower right is the one where you can type commands. You can move the cursor to the box above it (which contains responses from the program) by pressing the up arrow, and you can move the cursor back into the command line box by pressing return. -Type: read (, then drag your mol.txt file from its folder and drop it on the poy window. Poy will enter the path name to your file. Once it has been entered, type ), and hit enter. This will: Import
View Full Document
Unlocking...