[BioC] SPIA - changes between 1.4.0 and 1.60.

Sorin Draghici sorin at wayne.edu
Tue Sep 14 01:52:04 CEST 2010


I'm not sure whether you are interested in how the software was changed or how
those changes were documented.

For the former, as Adi said in his previous email, the first version ignored
nodes in the KEGG pathway that were associated with several genes.

For the latter, the link you included shows this for SPIA:
"updated pathway data by including interactions between genes and KEGG groups
that contain at least one gene"

Does this answer your question ?


On 9/13/2010 7:43 PM, Lavinia Gordon wrote:
>  Hi Sorin,
> thank you for your email.
> So to rephrase my question slightly, where does it say that nodes and links
> previously ignored are now considered in the analysis, comparing version 1.6 to
> 1.4?  The only document that I have found is here:
> http://fgc.lsi.umich.edu/cgi-bin/blosxom.cgi/2009/10/13
> which is no longer updated.
> If I look at the KEGG updates (http://www.genome.jp/kegg/docs/relnote.html), it
> would appear to suggest that no major changes have occurred in between the two
> SPIA versions.
> with regards
> Lavinia.
> On 14/09/2010 9:17 AM, Sorin Draghici wrote:
>> Lavinia,
>> The results are bound to be different because now: i) the KEGG pathways
>> themselves are different and ii) nodes and links previously ignored are now
>> considered in the analysis.
>> Regards,
>> Sorin
>> On 9/13/2010 7:13 PM, Lavinia Gordon wrote:
>>>   Hi Adi,
>>> Thank you for your answer, that is very helpful.
>>> Is there any document/site that indicates these changes?  I have seen changelogs
>>> before but wasn't able to find anything for SPIA.  The reason that I ask is that
>>> I have just re-run some old code and have quite different SPIA results between
>>> the old and new version.
>>> Thank you for your efforts in putting together this package.  It is extremely
>>> user-friendly and has generated some very nice results.
>>> with regards
>>> Lavinia.
>>> On 14/09/2010 1:02 AM, Tarca, Adi wrote:
>>>> Hi Lavinia,
>>>> This is a good question and I think other users would benefit from reading the
>>>> answer. Therefore it would be useful if you send your questions to the
>>>> bioconductor mailing list also.
>>>> I will give you the answer now anyhow.
>>>> The main change from SPIA 1.4.0 to 1.6.0 is the KEGG pathway data on which
>>>> SPIA works. KEGG pathways are updated periodically (genes are removed or
>>>> added, connections apear or are removed). In addition to this we also changed
>>>> the way we parse those raw pathway data. Initially only links between gene
>>>> type of nodes were considered. Now we consider also links between "group"
>>>> types of objects that include one or more genes.
>>>> Regards,
>>>> Adi
>>>> Adi Laurentiu Tarca, PhD
>>>> Assistant Professor (Research),
>>>> Department of Computer Science&   Center for Molecular Medicine and Genetics,
>>>> Wayne State University,
>>>> 3990 John R., Office 4809,
>>>> Detroit, Michigan 48201
>>>> Tel: 1-313-5775305
>>>> http://bioinformaticsprb.med.wayne.edu/tarca
>>>> -----Original Message-----
>>>> From: Lavinia Gordon [mailto:lavinia.gordon at mcri.edu.au]
>>>> Sent: Sunday, September 12, 2010 10:58 PM
>>>> To: Tarca, Adi
>>>> Subject: SPIA - changes between 1.4.0 and 1.60.
>>>>     Dear Dr Tarca,
>>>> My apologies for the simple question, as I am sure that this information must
>>>> be available, (at Bioconductor?), but I cannot find it.  I just wondered what
>>>> the changes were to the SPIA package between version 1.4.0 and 1.6.0?
>>>> with many thanks for your time,
>>>> Lavinia Gordon.
>>>> -- 
>>>> Senior Bioinformatics Officer
>>>> Murdoch Childrens Research Institute
>>>> Royal Children's Hospital
>>>> Flemington Road
>>>> Parkville
>>>> Victoria 3052
>>>> Australia
>>>> www.mcri.edu.au


Associate Professor
Director of the Intelligent Systems and Bioinformatics Laboratory
Department of Computer Science
Wayne State University
Room 431 State Hall,
5143 Cass Ave,
Detroit, MI 48202
Tel: 313-577-5484
Fax: 313-577-6868

More information about the Bioconductor mailing list