Exons:
Relevant bits of information for using the MUSC Shared Resource, 
the BioMolecular Computing Resource (BCR).
 
EXON_21 Running Des Higgons' CLUSTALW multiple sequence alignment
program.

The CLUSTALW program currently available is version 1.73.
CLUSATLW can read GCG multiple sequence files ( the output
files produced by GCG PILEUP program eg xxx.msf ), CLUSTAL 
format aligned sequences (eg yyy.aln) and Pearson 'fasta' format 
non-aligned files among other formats.

A useful procedure might be:

1) Construct file of files names you wish to align.
   This can be done with FASTA, STRINGSEARCH, LOOKUP
   WORDSEARCH, or manually.

2) Convert all sequences to 'fasta' format using the
   GCG program TOFASTA. TOFASTA understands the GCG
   "@" symbol so typing:
 
   tofasta @the_new.list

   Will produce an output file of concatenated, unaligned
   sequences which CLUSTALW can read.

3) Alternatively, retrieving files from internet sources
   with sequences in the "fasta" format.

4) Type "clustalw" and the program will start and display
   an initial menu allowing you to input sequences, proceed to
   multiple alignments, produce structurally constrained alignments
   or produce phylogenetic trees. At this point simply type '1' and
   enter the name of your GCG multiple alignment, the file of fasta 
   format sequences or the CLUSTAL format files.

5) There is a concise on-line help menu.

6) The output options include one very useful option, the ability
   to create output alignments in several formats. The most useful
   are: GCG MSF format, PHYLIP format (see exon 22) and of course
   the default CLUSTALW "aln" format.

7) Its possible to run CLUSTALW in batch mode but not all of the
   options of the interactive mode are accessible. Here is an example
   of a shell file for running CLUSTALW

   /usr/common/bcr/clustalw \
   /infile=xxxx.msf/align/outorder=aligned \
   /kimura/outfile=xxxx.phylip/output=PHYLIP

   The first line starts the program. The "\"
   character is a UNIX line continuation control character.
   The second line  defines the input file, tells the program
   to perform an alignment and to write the sequences out
   in the order of their alignment.
   The third line selects the kimura method for corrections
   for multiple mutations per site, names the output file
   and requests the PHYLIP output format 

8) If the name of the above shell file is "shell_clust" then to
   run the program in batch you would type "at 9:35" then
   hit return,
   then type "sh shell_clust" and hit return
   finally type control-d.

9) The happy result of which should be to start the alignment at
   9:30 and produce a phylip format multiple alignment.

10) Most of the time you will find that running clustalw is 
    quite efficiently accomplished via the interactive mode.