[BioC] KEGGgraph bug

[Marc Carlson]

Hi Fraser,

I believe that David fixed this bug a couple weeks ago.  Please use
biocLite() to update and let us know if there are any further issues.


  Marc





Fraser Sim wrote:
> Hi David,
>
> Thanks. That doesn’t solve the problem though, try to get the KEGG pathway
> ‘hsa03010’ using 
>
>   
>> getCategoryIndepKGMLurl(‘03010’,organism = ‘hsa’)
>>     
>
> It doesn’t produce a valid html link. It should find
> ‘ftp://ftp.genome.jp/pub/kegg/xml/kgml/non-metabolic/organisms/hsa/hsa03010.
> xml’ which is correct but because 'getKGMLurl' returns the invalid
> ‘ftp://ftp.genome.jp/pub/kegg/xml/kgml/non_metabolic/organisms/hsa/hsa03010.
> xml’, it then returns
> ‘ftp://ftp.genome.jp/pub/kegg/xml/kgml/metabolic/organisms/hsa/hsa03010.xml’
> which does not exist. 
>
> Note the substitution of "_" with "-". I’m pretty sure it’s a bug in
> getKGMLurl.
>
> Thanks,
> Fraser
>
> From: Jitao David Zhang [mailto:davidvonpku at gmail.com] 
> Sent: Monday, November 23, 2009 5:58 AM
> To: Fraser Sim
> Cc: bioc
> Subject: Re: [BioC] KEGGgraph bug
>
> Hi Fraser,
>
>   Thanks for the reporting.  The reason for this is that since April 2009
> the KEGG server saves the non-metabolic and metabolic pathways in two
> different sub-directory separatedly, and so far as I know there is no way to
> tell whether the pathway is metabolic alone from the pathway id. Therefore
> 'getKGMLurl' can NOT garantee to return the correct address. However,
> 'retrieveKGML' tries both metabolic and non-metabolic pathway directories,
> therefore it is able to download the file remotely.
>
>   To retrieve the URL for a pathway, use 'getCategoryIndepKGMLurl' function,
> which tries both non-metabolic and metabolic directories. 'getKGMLurl' will
> be hidden from the next release.
>
>   I hope my answer solves your question. 
>
>   Best wishes,
> David
> 2009/11/20 Fraser Sim <fjsim at buffalo.edu>
> I have found a problem with how 'getKGMLurl' constructs the URL for
> retrieving the KGML files. It looks like the '_' should be '-' in
> non_metabolic portion of the generated url. I am using the current release
> versions, R2.10.0 and KEGGgraph 1.2.0.
>
> Thanks,
> Fraser
>
>   
>> library(KEGGgraph)
>>     
> Loading required package: XML
> Loading required package: graph
>
> Attaching package: 'graph'
>
>
>        The following object(s) are masked from package:XML :
>
>         addNode
>
> Loading required package: Rgraphviz
> Loading required package: grid
>   
>> url <- getKGMLurl('04330', organism = "hsa")
>> url
>>     
> [1]
> "ftp://ftp.genome.jp/pub/kegg/xml/kgml/non_metabolic/organisms/hsa/hsa04330.
> xml"
>   
>> parseKGML(url)
>>     
> failed to load external entity
> "ftp://ftp.genome.jp/pub/kegg/xml/kgml/non_metabolic/organisms/hsa/hsa04330.
> xml"
> Error: 1: failed to load external entity
> "ftp://ftp.genome.jp/pub/kegg/xml/kgml/non_metabolic/organisms/hsa/hsa04330.
> xml"
>   
>> parseKGML(gsub("_","-",url))
>>     
> KEGG Pathway
> [ Title ]: Notch signaling pathway
> [ Name ]: path:hsa04330
> [ Organism ]: hsa
> [ Number ] :04330
> [ Image ] :http://www.genome.jp/kegg/pathway/hsa/hsa04330.png
> [ Link ] :http://www.genome.jp/kegg-bin/show_pathway?hsa04330
> ------------------------------------------------------------
> Statistics:
>        30 node(s)
>        16 edge(s)
>        0 reaction(s)
> ------------------------------------------------------------
>   
>> sessionInfo()
>>     
> R version 2.10.0 (2009-10-26)
> i386-pc-mingw32
>
> locale:
> [1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United
> States.1252    LC_MONETARY=English_United States.1252
> [4] LC_NUMERIC=C                           LC_TIME=English_United
> States.1252
>
> attached base packages:
> [1] grid      stats     graphics  grDevices datasets  utils     methods
> base
>
> other attached packages:
> [1] KEGGgraph_1.2.0  Rgraphviz_1.24.0 graph_1.24.0     XML_2.6-0
> rcom_2.2-1       rscproxy_1.3-1
>
> loaded via a namespace (and not attached):
> [1] tools_2.10.0
>   
>
> _______________________________________________
> 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
>
>
>
>