In article <3h7v3t$4cg at lyra.csx.cam.ac.uk> bcek1 at mole.bio.cam.ac.uk (Bernard Khoo (WCI)) writes:
>>I've done a 100 times bootstrap analysis on an 11 member dataset. However, when
>I try to run KITSCH on this dataset, specifying two user trees at the end as
>below, the program refuses to accept the user trees, complaining that either
>a parenthesis or semicolon must be missing.
>>[PROTDIST output x100]
>2
>(((scbrf,klbrf),cabrf),(((((hs2b,rat2b),xeno2b),dros2b),(sc2b,kl2b)),(pyrob,sul\
>fb)));
>(((scbrf,klbrf),cabrf),(((((hs2b,rat2b),xeno2b),dros2b),(pyrob,sulfb)),(sc2b,kl\
>2b)));
>>As far as I can see these trees are rooted and balanced. Not only that, they
>work as user trees if I append them to the end of the original (non-
>bootstrapped) dataset. Does anyone have any ideas as to what is happening here?
>Have I put the user trees in the wrong place in the file?
I think that, alas, you have to put them at the end of each of the 100
replicates of the Protdist output. This is not great user interface, I
admit. Why do you want to do user trees in a bootstrap run, though? Most
people don't need to.
It was complaining because it was reading into the next data set when it
was looking for a user tree.
-----
Joe Felsenstein, Dept. of Genetics, Univ. of Washington, Seattle, WA 98195
Internet: joe at genetics.washington.edu (IP No. 128.95.12.41)
