Central taxa analysis
gosink at u.washington.edu
Thu May 4 18:51:10 EST 1995
Is there a name for the type of analysis wherein a person:
1) Generates a distance matrix for a set of taxa, then
2) Adds the distances up for each organism, and finally
3) Finds the organism with the lowest total sum of distances.
Would this "least evolved" (whatever that means) organism be the closest
to the hypothetical ancestor of the taxa (under, say, a parsimony system)
- and hence retain the most ancestral features.
One immediate problem I see is that if several of the taxa are very
closely related to each other then the results would (naturally) be skewed
toward a member of that cluster.
The purpose of this question is that we have sequence data from a surface
epitope of a set of viral isolates and we want to find which isolate would
be best to use in producing immunoserum with the widest possible
crossreactivity. I know there are other, molecular-biology approaches but
we just wanted to try it from an evolutionary-biology angle.
More information about the Mol-evol