[Denovoassembler-devel] RE : RE : surveyor negative loaded sequence

[Sébastien B. <seb...@ul...>]

Done !


On 13 mars 2014 07:01, Maxime Déraspe [max...@gm...] wrote:
> À : Sébastien Boisvert; Maxime Deraspe
> Cc : den...@li...
> Objet : Re: [Denovoassembler-devel] RE : surveyor negative loaded sequence
> 
> Hi Seb,
> 
> I fixed an endless loop that happened in some cases in
> GenomeAssemblyReader when scaffolds were given in entry instead of contigs.
> 
> I also renamed the option to write the kmer matrix into a file, now
> "-write-kmer-matrix".
> Used to be "-run-kmer-matrix".
> 
> Please pull from :
> 
> https://github.com/Zorino/ray.git
> 
> fix-assembly-reader
> 
> Cheers,
> 
> Maxime
> 
> 
> On 03/12/2014 04:13 PM, Sébastien Boisvert wrote:
>> On 11 mars 2014 05:58, Maxime Deraspe [ma...@de...] wrote:
>>> À : Sébastien Boisvert
>>> Objet : surveyor negative loaded sequence
>>>
>>> Salut Seb,
>>>
>>> je me demandais comment est ce que le storekeeper pouvait loader une
>>> valeur négative de sequence.
>> The StoreKeeper actor is designed to only manage kmers in the distributed de Bruijn graph.
>> It does not load sequences from files directly. In fact, StoreKeeper actors
>> receive their payload from the aggregator (called Coa
>>
>>> /actors/1668 ->  loaded -1246000000 sequences
>>> /actors/1768 ->  loaded 1739000000 sequences
>>> /actors/1756 ->  loaded -1138000000 sequences
>>> /actors/1807 ->  loaded -2084000000 sequences
>>
>> If you look in the code, it is not a StoreKeeper actor that loads sequences from files and
>> that prints the "loaded XXX sequences" lines.
>>
>> [boiseb01@ls30 Surveyor]$ grep loaded *.cpp|grep sequences
>> GenomeAssemblyReader.cpp:       cout << " loaded " << m_loaded << " sequences" << endl;
>> GenomeGraphReader.cpp:                  cout << " loaded " << m_loaded << " sequences" << endl;
>>
>>>
>>> Est-ce que tu as une idée du problème qui peut survenir ?
>> It is in GenomeAssemblyReader.cpp or in GenomeGraphReader.cpp. GenomeGraphReader.cpp loads a graph file
>> and I tested it thoroughly.
>>
>> My guess would be that it is GenomeAssemblyReader.cpp, or code being used by this class.
>> More specifically, maybe you should look in SequenceKmerReader.cpp.
>>
>> In that file, SequenceKmerReader::hasAnotherKmer returns the value of m_hasKmerLeft.
>> The problem is presumably around that.
>>
>> What happens when your buffer has k symbols 'N' and you reached eof ?
>>
>> m_hasKmerLeft changes value on lines 67 and 118 in your code.
>>
>>
>> In your case, there is one thing to do to go forward:
>>
>>
>> Write a unit test for SequenceKmerReader.cpp.
>>
>>
>> You can add your unit test in https://github.com/sebhtml/Ray-TestSuite/tree/master/unit-tests if you want too.
>>
>>
>>
>>> Aussi je crois que j'ai mal compilé ray pour cette run, j'obtiens ces
>>> erreurs dans le stderr :
>>>
>>> Ray: /opt/pgi/linux86-64/11.8/libso/libnuma.so.1: no version information
>>> available (required by /opt/mpi/gcc/openmpi-1.6.4/lib64/libmpi.so.1)
>>>
>>> Maxime
>> ------------------------------------------------------------------------------
>> Learn Graph Databases - Download FREE O'Reilly Book
>> "Graph Databases" is the definitive new guide to graph databases and their
>> applications. Written by three acclaimed leaders in the field,
>> this first edition is now available. Download your free book today!
>> http://p.sf.net/sfu/13534_NeoTech
>> _______________________________________________
>> Denovoassembler-devel mailing list
>> Den...@li...
>> https://lists.sourceforge.net/lists/listinfo/denovoassembler-devel
>