Camelia, the Perl 6 bug

IRC log for #bioperl, 2011-01-26

| Channels | #bioperl index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:34 brandi joined #bioperl
00:35 brandi is now known as Guest17736
00:37 Guest17736 left #bioperl
00:42 ryan_ joined #bioperl
01:05 Vutral left #bioperl
01:21 Vutral joined #bioperl
01:25 ryan_ left #bioperl
02:15 Vutral left #bioperl
02:16 Vutral joined #bioperl
03:08 Vutral left #bioperl
04:26 randomstudent joined #bioperl
04:26 randomstudent hi
04:28 randomstudent left #bioperl
04:34 Vutral joined #bioperl
06:05 ank left #bioperl
06:52 Vutral left #bioperl
06:54 Vutral joined #bioperl
07:06 Vutral left #bioperl
09:22 kai morning folks
10:29 kai I'm afraid I don't fully understand the contig-assembly code in Bio/SeqIO/genbank.pm.. it's not possible to let it grab all the contig sequences from somewhere as far as I see, but it also doesn't look like it's possible to get that information to manually assemble this data either
12:29 maulikkamdar joined #bioperl
12:30 maulikkamdar Hello, Does Bioperl support vecscreen functions?
12:39 ank joined #bioperl
12:40 kai ah, I found an old script that I wrote sometime last year that does the assembly manually
13:02 maulikkamdar kai: An old script regarding vecscreen?? Can you please share a link?
13:26 maulikkamdar left #bioperl
13:44 kai no sorry. first of all you left, second of all it was about the stuff I talked a couple of hours ago
15:45 dbolser Use of uninitialized value in split at Bio/Root/Build.pm line 769, <STDIN> line 6.
15:46 dbolser kai: be fair, he waited nearly one whole half hour!
16:04 kblin dbolser: meh
16:04 dbolser [Wed Jan 26 16:02:14 2011] [error] [client 10.0.3.23] Math::BigInt: couldn't load specified math lib(s), fallback to Math::BigInt::Calc at /homes/www-potato/perl5/lib/perl5/Crypt/DH.pm line 6, referer: https://www.compbio.dundee.ac.uk/pot​ato/cgi-bin/gbrowse2/gbrowse/pot_qa/
16:04 dbolser [Wed Jan 26 16:02:13 2011] [error] [client 10.0.3.23] Undefined subroutine &Bio::DB::SeqFeature::Store::uncompress called at /homes/www-potato/perl5/lib/pe​rl5/Bio/DB/SeqFeature/Store.pm line 2531., referer: https://www.compbio.dundee.ac.uk/pot​ato/cgi-bin/gbrowse2/gbrowse/pot_qa/
16:04 dbolser lt latter is the showstopper
16:05 dbolser s/lt/the/
16:35 ank left #bioperl
16:35 ank joined #bioperl
16:56 ryan_ joined #bioperl
18:32 pyrimidine joined #bioperl
18:33 pyrimidine kai: re: contig assembly code, I think if you retrieve the sequence via Bio::DB::GenBank the underlying code does some magic and assembles the sequence for you.
18:37 perl_splut dbolser, looks like that was removed at some point from live
18:37 perl_splut unfortunately the deobfuscator only works against live and not specific versions to see when it was moved/removed
18:40 dbolser perl_splut: I bugged the list, didn't find any similar problems though...
18:41 dbolser lets see if someone replies
18:41 perl_splut did you check the gbrowse docs for what version of BioPerl they want you to use?
18:41 dbolser oh
18:42 blarney joined #bioperl
18:48 pyrimidine dbolser: my guess is they want v1.6.1 off CPAN unless you are installing using the net_install script (which I'm not sure works with GB2)
18:53 dbolser how do I get that using git? is it on the bioperl git page?
18:53 dbolser if so, I'll just go read there instead of botheing you
18:54 dbolser I'm asking because you wrote most of that page, and it 'll take me at least 30 s to scan it
18:54 perl_splut don't use git
18:54 perl_splut use cpan
18:55 pyrimidine If you want the latest cutting edge code: http://gmod.svn.sourceforge.net/viewvc/gmod/Gener​ic-Genome-Browser/trunk/bin/gbrowse_netinstall.pl
18:55 pyrimidine (-d option I think)
18:55 pyrimidine Oh wait, that's a dead link
18:58 pyrimidine https://gmod.svn.sourceforge.net/svnroot/gmod/Gene​ric-Genome-Browser/trunk/bin/gbrowse_netinstall.pl
19:15 blarney anybody know of a way of taking a list of GO terms that are enriched and grouping them using GO Slim?
19:15 blarney Perl, Python script etc.
19:16 blarney there is go-perl, but it appears to rely on being given a list of genes, rather than GO terms
19:19 pyrimidine blarney: you mean something like this? http://search.cpan.org/~cmungall/go-​perl-0.13/scripts/go-export-graph.pl
19:20 pyrimidine blarney: of course, R/BioC is good for this as well
19:46 CIA-66 left #bioperl
20:01 blarney pyrimidine: thanks
20:10 CIA-67 joined #bioperl
20:14 kblin pyrimidine: ah, that's pretty much what I wrote myself as well
20:14 kblin pyrimidine: just that my code works on .embl files :)
20:15 pyrimidine kblin: ok.  one would think nice functions like that should be parceled out somewhere more generically useful in bioperl :)
20:18 kblin pyrimidine: yeah, I've almost misplaced that script myself, I'd rather keep a backup on bioperl-live.git
20:18 kblin just need to clean this up a bit and see if I can make it work for genbank files as well
20:19 kblin I take a script like that would go to Bio/Tools?
20:22 pyrimidine hmm, not sure.  Thinking it should be something possibly exportable into a module's namespace, so one could use it like:
20:22 pyrimidine parse_contig_string($str)
20:22 pyrimidine via 'use Bio::Tools::Foo qw(parse_contig_string);'
20:22 pyrimidine or somesuch
20:24 kblin should be possible
20:25 pyrimidine we sort of have this already with some modules in the Bio::Align namespace
20:25 pyrimidine might be nice to have this in Bio::Tools as you say
20:25 pyrimidine so, something like Bio::Tools::SeqTools
20:27 pyrimidine There is a Bio::SeqUtils, but looks like everything is a method (object-related) instead of exportable.  Pity.
20:38 blarney pyrimidine: actually, what I really need to do is group similar GO terms, so it doesn't necessarily have be using GO Slim
20:41 pyrimidine blarney: The hypergeometric analyses packages in BioC do this, IIRC
20:47 pyrimidine blarney: these generally start from a gene list IIRC
20:48 blarney blarney: right, and that's the problem, I already have the enriched terms using the funcassociate tool, I just want to group them (since there are so many)
20:48 blarney funcassociate doesn't have a way of using GO Slim instead of full GO
20:49 pyrimidine how about a simple binning procedure for each GO category; keep track of number of times GO terms pop up after crawling up the GO hierarchy for all your terms
20:49 pyrimidine quick and dirty
21:00 blarney using the raw obo file or something?
21:13 pyrimidine blarney: Bio::Ontology reads obo format; you could try using that and generate your own script
21:13 pyrimidine http://search.cpan.org/~cjfield​s/BioPerl/Bio/OntologyIO/obo.pm
21:17 pyrimidine This would return a Bio::Ontology::Ontology, where you can do things like ask it: my $term = $ontology->find_terms(-identifier => 'foo')', then recursively crawl the tree using 'my @ancestor_terms = $ontology->get_ancestor_terms($term)'
21:18 pyrimidine http://search.cpan.org/~cjfields/​BioPerl/Bio/Ontology/Ontology.pm
21:18 pyrimidine each of those 'terms' is a Bio::Ontology::Term object, so you can ask for its name, etc
21:19 pyrimidine blarney: not sure if you followed that ^^^^^
21:19 blarney thanks, I mostly use biopython, but that channel seems so quiet, I do have bioperl installed as well though
21:20 pyrimidine they may have ontology tools as well, not sure
21:21 blarney not quite ready for prime-time it seems: http://www.biopython.org/w/index.php​?title=Gene_Ontology&amp;redirect=no
21:23 blarney I'd want it to stop, when it reached a GO slim term, I guess, otherwise it will always get to the top of the tree
21:26 pyrimidine one could do that; just need to iterate through the ancestor terms and stopping when you see a term that exists in a blacklist of unwanted terms
21:38 pyrimidine left #bioperl
21:44 CIA-67 bioperl-live: Johan Viklund master * r68979de / (924 files in 164 dirs): Merge branch 'master' of git://github.com/bioperl/bioperl-live - http://bit.ly/fKhQBg
21:44 CIA-67 bioperl-live: Johan Viklund master * rac84449 / (Bio/SeqIO/swiss.pm t/SeqIO/swiss.t): [Bio::SeqIO::swiss] Fixed bug where write_seq() were generating non-standard swiss and added tests for this - http://bit.ly/e9pKTZ
22:07 dukeleto left #bioperl
22:07 dukeleto joined #bioperl
23:20 blarney left #bioperl
23:42 melic left #bioperl
23:49 melic joined #bioperl

| Channels | #bioperl index | Today | | Search | Google Search | Plain-Text | summary