Question Regarding Bootstrap

James McInerney james.o.mcinerney at
Tue Oct 19 10:25:11 EST 1999

> Second, the bootstrap values will not change for a maximum parsimony
> analysis when you add invariable characters; although they can change a
> little using some other methods, like maximum likelihood.  For maximum
> likelihood, the result is not predictable; sometimes the bootstrap values
> will increase, sometimes they will decrease.


Just to take issue with one point in your otherwise impeccable answer. 
Bootstrap values can be affected by the inclusion or exclusion of
constant characters for a parsimony analysis.  This will affect the
sampling process.  If you consider this alignment:

species1    acgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgttag
species2    acgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtaga
species3    acgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtaga
species4    acgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgtacgttag

There are three parsimony-informative sites, all supporting the same
topology.  By removing the 60 constant sites you will get a bootstrap
value for the internal branch that is 100%.  If you include the constant
sites, you will only get a bootstrap value of 95.4% (you should do about
one million replicates to get a reasonably accurate estimate of the
b'strap value).

So, the inclusion or exclusion of these sites can affect the bootstrap
values.  There are probably better cases, this is just a quick alignment
I put together.

Kindest regards,


More information about the Mol-evol mailing list