[BioC] error in plotKEGGGraph

Pierre Neuvial pierre at stat.berkeley.edu
Tue Sep 21 23:54:47 CEST 2010 

Hi David,

Has there been any follow-up on this ? I bumped into the same error as Robert.

You asked for the file he used.  If you run the code sent by Robert,
you will get

> keggID <- "05213"
> tmpXML <- "testKGML1.xml"
> retrieveKGML(pathwayid=keggID, organism='mmu', destfile=tmpXML, method="wget")
--2010-09-21 12:23:04--
ftp://ftp.genome.jp/pub/kegg/xml/kgml/non-metabolic/organisms/mmu/mmu05213.xml
          => `/var/folders/R7/R7TMUh6WGeSJE0LtxqwHWk+++TI/-Tmp-//RtmpmGNCWB/file1eaf6b0f'
Resolving ftp.genome.jp... 133.103.100.184
Connecting to ftp.genome.jp|133.103.100.184|:21... connected.
Logging in as anonymous ... Logged in!
==> SYST ... done.    ==> PWD ... done.
==> TYPE I ... done.  ==> CWD
/pub/kegg/xml/kgml/non-metabolic/organisms/mmu ... done.
==> SIZE mmu05213.xml ... 16729
==> PASV ... done.    ==> RETR mmu05213.xml ... done.
Length: 16729 (16K)

100%[======================================>] 16,729      38.4K/s   in 0.4s

2010-09-21 12:23:07 (38.4 KB/s) -
`/var/folders/R7/R7TMUh6WGeSJE0LtxqwHWk+++TI/-Tmp-//RtmpmGNCWB/file1eaf6b0f'
saved [16729]

--2010-09-21 12:23:07--
ftp://ftp.genome.jp/pub/kegg/xml/kgml/non-metabolic/organisms/mmu/mmu05213.xml
          => `testKGML1.xml'
Resolving ftp.genome.jp... 133.103.100.184
Connecting to ftp.genome.jp|133.103.100.184|:21... connected.
Logging in as anonymous ... Logged in!
==> SYST ... done.    ==> PWD ... done.
==> TYPE I ... done.  ==> CWD
/pub/kegg/xml/kgml/non-metabolic/organisms/mmu ... done.
==> SIZE mmu05213.xml ... 16729
==> PASV ... done.    ==> RETR mmu05213.xml ... done.
Length: 16729 (16K)

100%[======================================>] 16,729      44.2K/s   in 0.4s

2010-09-21 12:23:09 (44.2 KB/s) - `testKGML1.xml' saved [16729]

That should be enough for you to reproduce the error.  If not, the file is here:

ftp://ftp.genome.jp/pub/kegg/xml/kgml/non-metabolic/organisms/mmu/mmu05213.xml

I've investigated this issue a bit more. The problem comes from the
fact that this KEGG pathway (as well as a fair proportion of KEGG
pathways) has at least one edge whose label is not understood by
KEGGgraph, in this case (quoted from the above KGML file):

   <relation entry1="20" entry2="25" type="PCrel">
       <subtype name="missing interaction" value="-/-"/>
   </relation>

The types of interaction known by KEGGgraph are retrieved from a data
object from the package:

> data(KEGGEdgeSubtype, package="KEGGgraph")
> as.character(KEGGEdgeSubtype[["name"]])
 [1] "compound"            "hidden compound"     "activation"
 [4] "inhibition"          "expression"          "repression"
 [7] "indirect"            "indirect effect"     "state change"
[10] "binding/association" "dissociation"        "phosphorylation"
[13] "dephosphorylation"   "glycosylation"       "ubiquitination"
[16] "methylation"

Because "missing interaction" is not known, the above error occurs.
An easy fix for this problem is to add a line with name=="missing
interaction" to this data.frame, and re-build the package.

I've also encountered another instance of the same problem:

> keggID <- "05213"
> tmpXML <- "testKGML2.xml"
> retrieveKGML(pathwayid=keggID, organism='mmu', destfile=tmpXML, method="wget")
--2010-09-21 13:56:37--
ftp://ftp.genome.jp/pub/kegg/xml/kgml/non-metabolic/organisms/mmu/mmu05213.xml
          => `/var/folders/R7/R7TMUh6WGeSJE0LtxqwHWk+++TI/-Tmp-//RtmpmGNCWB/file13f877e9'
Resolving ftp.genome.jp... 133.103.100.184
Connecting to ftp.genome.jp|133.103.100.184|:21... connected.
Logging in as anonymous ... Logged in!
==> SYST ... done.    ==> PWD ... done.
==> TYPE I ... done.  ==> CWD
/pub/kegg/xml/kgml/non-metabolic/organisms/mmu ... done.
==> SIZE mmu05213.xml ... 16729
==> PASV ... done.    ==> RETR mmu05213.xml ... done.
Length: 16729 (16K)

100%[======================================>] 16,729      41.4K/s   in 0.4s

2010-09-21 13:56:39 (41.4 KB/s) -
`/var/folders/R7/R7TMUh6WGeSJE0LtxqwHWk+++TI/-Tmp-//RtmpmGNCWB/file13f877e9'
saved [16729]

--2010-09-21 13:56:39--
ftp://ftp.genome.jp/pub/kegg/xml/kgml/non-metabolic/organisms/mmu/mmu05213.xml
          => `testKGML2.xml'
Resolving ftp.genome.jp... 133.103.100.184
Connecting to ftp.genome.jp|133.103.100.184|:21... connected.
Logging in as anonymous ... Logged in!
==> SYST ... done.    ==> PWD ... done.
==> TYPE I ... done.  ==> CWD
/pub/kegg/xml/kgml/non-metabolic/organisms/mmu ... done.
==> SIZE mmu05213.xml ... 16729
==> PASV ... done.    ==> RETR mmu05213.xml ... done.
Length: 16729 (16K)

100%[======================================>] 16,729      58.8K/s   in 0.3s

2010-09-21 13:56:42 (58.8 KB/s) - `testKGML2.xml' saved [16729]

> tmpGraph2 <- parseKGML2Graph(tmpXML, genesOnly=FALSE)
> plotKEGGgraph(tmpGraph2)
Error in .subtypeDisplay(object) :
 Given subtype 'ubiquination' is not found!

This actually comes from a typo in KEGG entries: ubiquitination has
been misspelled as "ubiquination" in
ftp://ftp.genome.jp/pub/kegg/xml/kgml/non-metabolic/organisms/mmu/mmu05213.xml
(and a bunch of other KEGG files too).

Again, this can be fixed by adding a line with name=="ubiquination" to
KEGGEdgeSubtype.
I've done this locally: edited the file
KEGGgraph/data/KEGGEdgeSubtype.csv, recompiled the KEGGgraph package,
and it works with no error.

David, could you do this kind of fix ?

More generally, it would of course be much better if the
.subtypeDisplay function did not throw an *error* but a *warning* when
encountering unknown interaction names.

Hope this helps,

Pierre

R version 2.11.1 Patched (2010-09-16 r52946)
Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)

locale:
[1] en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] grid      stats     graphics  grDevices datasets  utils     methods
[8] base

other attached packages:
[1] KEGGgraph_1.5.0 Rgraphviz_1.26.0  graph_1.26.0      XML_2.8-1
[5] R.utils_1.5.2     R.oo_1.7.3        R.methodsS3_1.2.0 fortunes_1.3-7

loaded via a namespace (and not attached):
[1] tools_2.11.1

On Wed, Jul 21, 2010 at 2:31 PM, Jitao David Zhang
<davidvonpku at gmail.com> wrote:
> Hi dear Robert,
>
>   Thanks for reporting! Would you please send the file you used? I will try
> to figure out what went wrong.
>
>  Best wishes,
> David
>
> 2010/7/7 Robert M. Flight <rflight79 at gmail.com>
>
>> Hi All,
>>
>> I'm playing around with KEGGGraph to examine some pathways, and I am
>> running into some errors.
>>
>> When I try to plot some pathways, I get the error: "Error in
>> .subtypeDisplay(object) :  Given subtype 'missing interaction' is not
>> found!"
>>
>> Here is an example with one pathway that gives the error and another
>> that does not:
>>
>> library(KEGGgraph)
>>
>> # this one gives the error
>> keggID <- "05213"
>> tmpXML <- "testKGML1.xml"
>> retrieveKGML(pathwayid=keggID, organism='mmu', destfile=tmpXML,
>> method="wget")
>> tmpGraph1 <- parseKGML2Graph(tmpXML, genesOnly=FALSE)
>> plotKEGGgraph(tmpGraph1)
>>
>> # this one does not give an error, and spits out a very complicated graph
>> keggID <- "04810"
>> tmpXML <- "testKGML2.xml"
>> retrieveKGML(pathwayid=keggID, organism='mmu', destfile=tmpXML,
>> method="wget")
>> tmpGraph2 <- parseKGML2Graph(tmpXML, genesOnly=FALSE)
>> plotKEGGgraph(tmpGraph2)
>>
>> Any help on why this error is cropping up would be appreciated.
>>
>> R version 2.11.1 (2010-05-31)
>> i386-pc-mingw32
>>
>> locale:
>> [1] LC_COLLATE=English_United States.1252
>> [2] LC_CTYPE=English_United States.1252
>> [3] LC_MONETARY=English_United States.1252
>> [4] LC_NUMERIC=C
>> [5] LC_TIME=English_United States.1252
>>
>> attached base packages:
>> [1] grid      stats     graphics  grDevices utils     datasets  methods
>> [8] base
>>
>> other attached packages:
>> [1] KEGGgraph_1.4.0  Rgraphviz_1.26.0 graph_1.26.0     XML_3.1-0
>>
>> loaded via a namespace (and not attached):
>> [1] tools_2.11.1
>>
>> Thanks in advance,
>>
>> -Robert
>>
>> Robert M. Flight, Ph.D.
>> Bioinformatics and Biomedical Computing Laboratory
>> University of Louisville
>> Louisville, KY
>>
>> PH 502-852-0467
>> EM robert.flight at louisville.edu
>> EM rflight79 at gmail.com
>>
>> Williams and Holland's Law:
>>       If enough data is collected, anything may be proven by
>> statistical methods.
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>
>
>
>
> --
> Jitao David Zhang
> Pre-doc staff of Computational Biology and Biostatistics
> Division of Molecular Genome Analysis
> DKFZ, Heidelberg D-69120, Germany
>
> http://www.NextBioMotif.com/
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>

More information about the Bioconductor mailing list