[Phylobase-devl] characters not being read
François Michonneau
francois.michonneau at gmail.com
Sun Aug 8 10:12:22 CEST 2010
Hi,
I feel there are 2 different issues here.
For the tip labels, I feel it would be best to keep the labels as they
are returned from NCL to stay consistant.
For the CHARSTATELABELS we can add the option check.names to
readNexus.
Cheers,
-- François
On Thu, 2010-08-05 at 15:37 -0400, Conrad Stack wrote:
> What about, since phylobase requires ape, importing the checkLabel
> function from ape and filtering all tip label through that function?
>
>
> -conrad
>
> On Thu, Aug 5, 2010 at 10:00 AM, Ben Bolker <bolker at ufl.edu> wrote:
> François Michonneau wrote:
> > Hi again Conrad,
> >
> > In the file you sent me there were no state labels (just
> > charstatelabels), so the warning is normal.
> >
> > For the underscore issue, there isn't much we can do
> about. NCL
> > converts underscores to spaces because there is no way to
> > differentiate them in newick strings for instance. Then when
> the name
> > is assigned to the column name in R, the space is converted
> to a dot.
> >
> > Cheers,
> > -- François
>
>
> If we wanted to we could (a) convert spaces back to
> underscores
> ourselves after NCL import (b) disable conversion of spaces to
> dots by
> using check.names=FALSE in the data.frame() call.
>
> I don't know whether either of these is useful in general,
> or if we
> just need to 'fix' this in the documentation.
>
> Opinions?
>
> Ben
>
>
> >
> > On Wed, Aug 4, 2010 at 22:31, Conrad Stack <stack at psu.edu>
> wrote:
> >> Looks great. Two things: it looks like the warning
> message about 'state
> >> labels' fires even when labels are found (only when
> DATATYPE=STANDARD) and,
> >> perhaps relatedly, underscores in CHARSTATELABELS are
> automatically
> >> converted to periods.
> >> Thanks!
> >> -Conrad
> >>
> >>
> >>
> >> On Wed, Aug 4, 2010 at 12:53 PM, François Michonneau
> >> <francois.michonneau at gmail.com> wrote:
> >>> Hi,
> >>>
> >>> My last commit should fix the issue Conrad mentioned.
> >>>
> >>> The problem came from the fact that the nexus file didn't
> have state
> >>> labels and the default option in readNexus was to return
> them. It
> >>> ended up attaching empty labels to the the 0 and 1 which
> resulted in
> >>> the empty column observed. I modified readNexus to (1)
> return a
> >>> warning when state labels are missing and return.labels is
> TRUE which
> >>> leads to ignoring the return.labels option.
> >>>
> >>> Let me know if you find any other bugs.
> >>>
> >>> Cheers,
> >>> -- François
> >>>
> >>> On Tue, Aug 3, 2010 at 17:42, Conrad Stack <stack at psu.edu>
> wrote:
> >>>> I just upgraded to phylobase 0.6.1 and found that the
> data in my
> >>>> CHARACTERS
> >>>> blocks are no longer being read in (they show up as
> blanks in the
> >>>> returned
> >>>> phylo4d object).
> >>>> this is from the geospiza.nex dataset:
> >>>> BEGIN CHARACTERS;
> >>>> DIMENSIONS NCHAR=1;
> >>>> FORMAT DATATYPE = STANDARD GAP = - MISSING = ? SYMBOLS =
> " 0 1";
> >>>> MATRIX
> >>>> fuliginosa 1
> >>>> fortis 1
> >>>> magnirostris 0
> >>>> conirostris 0
> >>>> scandens 0
> >>>> difficilis 1
> >>>> pallida 0
> >>>> parvulus 0
> >>>> psittacula 0
> >>>> pauper 0
> >>>> Platyspiza 0
> >>>> fusca 0
> >>>> Pinaroloxias 0
> >>>> ;
> >>>> END;
> >>>> When it reads in, it shows up like this:
> >>>> readNexus("geospiza.nex")
> >>>> label node ancestor edge.length node.type
> standard_char
> >>>> 1 fuliginosa 1 22 0.05500 tip
> >>>> 2 fortis 2 22 0.05500 tip
> >>>> 3 magnirostris 3 21 0.11000 tip
> >>>> 4 conirostris 4 20 0.18333 tip
> >>>> 5 scandens 5 19 0.19250 tip
> >>>> 6 difficilis 6 18 0.22800 tip
> >>>> 7 pallida 7 23 0.08667 tip
> >>>> 8 parvulus 8 25 0.02000 tip
> >>>> 9 psittacula 9 25 0.02000 tip
> >>>> 10 pauper 10 24 0.03500 tip
> >>>> 11 Platyspiza 11 16 0.46550 tip
> >>>> 12 fusca 12 15 0.53409 tip
> >>>> 13 Pinaroloxias 13 14 0.58333 tip
> >>>> 14 <NA> 14 0 0.29744 root
> <NA>
> >>>> 15 <NA> 15 14 0.04924 internal
> <NA>
> >>>> 16 <NA> 16 15 0.06859 internal
> <NA>
> >>>> .......cont....
> >>>> Any thoughts? (I'm using windows 7 32-bit with R 2.11,
> but the error
> >>>> occurs
> >>>> on my Mac leopard install as well.
> >>>>
> >>>>
> >>>> --
> >>>> Conrad Stack
> >>>> -----------------------
> >>>> PSU Department of Biology
> >>>> 208 Mueller Lab
> >>>> University Park, PA 16802
> >>>> cell: 814.409.8310
> >>>> email: conrad.stack at gmail.com
> >>>>
> >>>> _______________________________________________
> >>>> Phylobase-devl mailing list
> >>>> Phylobase-devl at lists.r-forge.r-project.org
> >>>>
> >>>>
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/phylobase-devl
> >>>>
> >>>>
> >>
> >>
> >> --
> >> Conrad Stack
> >> -----------------------
> >> PSU Department of Biology
> >> 208 Mueller Lab
> >> University Park, PA 16802
> >> cell: 814.409.8310
> >> email: conrad.stack at gmail.com
> >>
> > _______________________________________________
> > Phylobase-devl mailing list
> > Phylobase-devl at lists.r-forge.r-project.org
> >
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/phylobase-devl
>
>
> --
>
> Ben Bolker
> *** NEW E-MAIL ADDRESSES:
> *** bbolker at gmail.com , bolker at mcmaster.ca
> http://www.math.mcmaster.ca/~bolker
> GPG key:
> http://www.math.mcmaster.ca/~bolker/benbolker-publickey.asc
>
>
>
>
> --
> Conrad Stack
> -----------------------
> PSU Department of Biology
> 208 Mueller Lab
> University Park, PA 16802
> cell: 814.409.8310
> email: conrad.stack at gmail.com
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part
URL: <http://lists.r-forge.r-project.org/pipermail/phylobase-devl/attachments/20100808/9f933edc/attachment.pgp>
More information about the Phylobase-devl
mailing list