[Biophp-dev] hmmpfam parser

yvan biophp-dev@bioinformatics.org
Thu, 12 Aug 2004 15:04:04 +1000

Hi guys,

Sorry for the delay.

Your parser is more finished then mine, furthermore my script use the  
expat library which i guess make it more tricky to use. There is no  
reason to depose it unless you want to have a look at. I am afraid now  
it is more a script to parse a blast xml output and fill a database.  
For the hmmpfam parser, it is very simple and a little buggy. It  only  
parse file produced by hmmpfam, but you can have a look if you want. I  
don't have time to make it PEARish. What it does. take a file  
containing one or more hmmpfam outputs  and return a array of the kind:  
array[sequence_name][domains][domains_data]. Being listed as a  
developer, seems a little be to much as i don't know how much i could  
contribute. So anyone who want the script let me know.


On 27/07/2004, at 7:09 AM, S Clark wrote:

> Argh - sorry about the delay in posting - works insane right now...
> There actually is already a BLAST xml parser in CVS, though if yours
> is better there's no reason we can't change over (for that matter
> there's not really any reason we couldn't compare the two and combine  
> them).
> If you want to have a look, you can see it in CVS at:
> https://bioinformatics.org/cgi-bin/cvsweb.cgi/biophp/sean/semi- 
> official/ncbi_blast_parse_xml.inc.php
> and the base xml parser from which this class is derived is at:
> https://bioinformatics.org/cgi-bin/cvsweb.cgi/biophp/sean/semi- 
> official/support_modules/parser_xml.inc.php
> Comments and suggestions welcome, as usual.  (I just noticed that  
> although the
> frontend class got copied over to the more official "genephp"  
> subdirectory,
> the parsers and NCBI-BLAST frontend didn't, apparently - will fix that
> later...)
> Note that the "parsers" actually interpret the results and contain  
> methods
> for fetching the individual result fields as needed.  The whole result  
> in
> the form of an associative array (or "hash table" or whatever you want  
> to call
> it) is stored by the parser.
> If you want to contribute code, just let me know what your login is
> at bioinformatics.org and I can add you to the developers list.
> Thanks!
> Sean
> On Friday 09 July 2004 10:42 am, Nico Stuurman wrote:
>> Hi Yvan,
>> You'll first need an account at bioinformatics.org and Sean has to add
>> you as a developer.
>> As for the code itself, we try to adhere the PEAR coding standards
>> (http://pear.php.net/manual/en/standards.php).  Not all the code there
>> has been PEARified yet, but you can look at biophp/genephp/seq.inc.php
>> as an example.  Try to provide ample documentation and references to
>> specific algorithms etc. if applicable.
>> I am not sure where in the structure your parsers would fit (Sean, do
>> you have a good idea about this?).  biophp/genephp/parsers is not a
>> good idea since these are all parsers for sequence file formats (I
>> guess we should rename parsers to importers).
>> biophp/genephp/interfaces and biophp/genephp/utilities seem equally
>> applicable right now.
>> In the long run we should get some kind of datastructure holding  
>> output
>> from blast/fasta like services.  It would be great if you and Sean
>> coudl think about that.
>> Hope this helps,
>> Nico
>>> Just wondering what are the requirements for submiting codes to
>>> Biophp? I wrote two parsers for hmmpfam and blast xml outputs, but am
>>> afraid they are far from being useable in the context of biophp.
>>> Please, could someone explain me the DO and DONT?
>>> Regards
>>> Yvan
>>> _______________________________________________
>>> Biophp-dev mailing list
>>> Biophp-dev@bioinformatics.org
>>> https://bioinformatics.org/mailman/listinfo/biophp-dev
>> _______________________________________________
>> Biophp-dev mailing list
>> Biophp-dev@bioinformatics.org
>> https://bioinformatics.org/mailman/listinfo/biophp-dev