Oscar:
Oscar Franzén wrote on 17/03/2010 17:55:
Dear all,
I'm trying to use the boot.phylo function in the ape package to get
bootstrap values for my NJ-tree which is based on a clustalw alignment.
Basically what I'm doing is this:
library(ape)
library(seqinr)
an = read.alignment("file.aln", format="clustal")
an.dist = dist.alignment(an)
an.nj = nj(an.dist)
qwe = as.DNAbin(an)
any(is.na(an.dist))
[1] FALSE
If I call boot.phylo this happens:
You should use in boot.phylo the same function used to estimate the tree
'an.nj', but that's not what you did below.
Besides, I suspect that your alignment is of AA, not DNA. If true, you
can't use dist.dna.
Emmanuel
for (i in 1:5) boot.phylo(an.nj, qwe, function(xx) nj(dist.dna(xx)))
Error in nj(dist.dna(xx)) :
missing values are not allowed in the distance matrix
boot.phylo(an.nj, qwe, function(xx) nj(dist.dna(xx)))
*** caught segfault ***
address 0x9c11039, cause 'memory not mapped'
Traceback:
1: .C("dist_dna", x, n, s, imod, double(Ndist), BF,
as.integer(pairwise.deletion), as.integer(variance), var,
as.integer(gamma), alpha, DUP = FALSE, NAOK = TRUE, PACKAGE = "ape")
2: dist.dna(xx)
3: nj(dist.dna(xx))
4: FUN(x[, boot.samp])
5: boot.phylo(an.nj, qwe, function(xx) nj(dist.dna(xx)))
Possible actions:
1: abort (with core dump, if enabled)
2: normal R exit
3: exit R without saving workspace
4: exit R saving workspace
Selection:
But the weird thing is that it sometimes works:
boot.phylo(an.nj, qwe, function(xx) nj(dist.dna(xx)))
Error in nj(dist.dna(xx)) :
missing values are not allowed in the distance matrix
boot.phylo(an.nj, qwe, function(xx) nj(dist.dna(xx)))
[1] 100 0 0 0 11 0 0 0 0 0 0 0 0 0 0 18 26
0
[19] 0 7 0 0 0 0 0 16 0 1 0 0 16 4 0 7 0
0
[37] 0 3 0 8 21 6 0 0 0 18 0 0 0 3 0 0 13
13
[55] 0 16 3 13 48 66 0 1 0 0 0 15 0 3 25 0 4
0
[73] 0 19 11 0 33 1 28 0 8 15 1 15 29 34 0 15 8
0
[91] 29 4 3 11 0 7 1 16 3 6 6 13 10 19 25
Can this be a bug in R/boot.phylo or could the error be located in my
dataset? My dataset is a gene family consisting of around 100 members.
I just upgraded to R v2.10.1, and the problem remains.
Thanks in advance,
Oscar
[[alternative HTML version deleted]]
_______________________________________________
R-sig-phylo mailing list
R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
--
Emmanuel Paradis
IRD, Montpellier, France
ph: +33 (0)4 67 16 64 47
fax: +33 (0)4 67 16 64 40
http://ape.mpl.ird.fr/
_______________________________________________
R-sig-phylo mailing list
R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo