Percent identity for numerous comparisons
James O. McInerney
j.mcinerney at nhm.ac.uk
Tue May 5 10:04:57 EST 1998
Why not draw a phylogenetic tree? This tree is a little cartoon, representing
the relationships between the sequences. You can compute pair-wise distances
between your sequences and then use a clustering method (or some of the more
sophisticated 'tree-searching' methods) to produce a dendrogram of the
inferred relationships. The distances along the branches from one sequence to
another is an indication of the amount of sequence difference between the two.
One of the most popular suites of programs for performing these kinds of
analyses is the PHYLIP package with is available through the GCG package or
can be obtained from:
The whole idea of drawing a phylogenetic tree is to graphically represent the
relationships between sequences. If the tree is drawn from an initial
pair-wise distance matrix, then it will attempt to graphically represent this matrix.
Hope this helps.
Elisa Eshbaugh wrote:
> Hello all and thanks to those who can make a suggestion...
> I am calculating %identity (amino acid and nucleotide) among 43
> different taxa representing 4 subfamilies of a gene family. This means
> that by the time I have done every possible pairwise comparison there
> will be hundreds. There's really no problem in doing this, but it kind
> of eliminates the potential for putting all the information in a nice
> little table. If I want to extract as much information as possible, but
> eliminate any less than necessary work, what might be some criteria for
> eliminating some of the comparisons?
> Thanks again.
James O. McInerney email: J.mcinerney at nhm.ac.uk
Molec. Biol. Comput. Officer, phone: +44 171 938 9163
Department of Zoology, Fax: +44 171 938 9158
The Natural History Museum,
London SW7 5BD.
More information about the Mol-evol