Camelia, the Perl 6 bug

IRC log for #bioperl, 2011-03-15

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

All times shown according to UTC.

Time Nick Message
01:45 bag_ left #bioperl
04:41 zenman_ joined #bioperl
06:19 zenman_ left #bioperl
07:19 bag_ joined #bioperl
08:37 bag_ left #bioperl
09:54 dbolser left #bioperl
09:57 dbolser joined #bioperl
11:11 ank left #bioperl
12:08 kai left #bioperl
12:32 zenman_ joined #bioperl
12:46 zenman_ left #bioperl
12:53 kai joined #bioperl
12:53 kai left #bioperl
12:53 kai joined #bioperl
13:28 reece left #bioperl
14:45 pyrimidine Amazing: "I can't use Bio::DB::GenBank to pull 1 million seqs from NCBI; why doesn't that work"
14:45 pyrimidine GAHH!
14:46 pyrimidine at the moment I don't care if this is logged; bp users should realize this is a very stupid thing to do
15:01 dbolser but I want 1G
15:08 reecehart joined #bioperl
15:25 ank joined #bioperl
15:25 cassj joined #bioperl
15:33 deafferret the htc evo has more Gs
15:42 dbolser can I install bioperl on that?
15:46 deafferret if you root kit it, yes
15:46 deafferret so i've heard
15:50 dbolser nice
15:57 cassj left #bioperl
16:58 kyanardag_ joined #bioperl
17:09 reecehart1 joined #bioperl
17:10 reecehart left #bioperl
17:16 sheenams joined #bioperl
17:22 kyanardag_ left #bioperl
17:45 reecehart1 left #bioperl
17:48 reecehart joined #bioperl
18:29 andrei_ joined #bioperl
18:29 andrei_ hey guys I have a very noobish question
18:30 andrei_ it's for the code here https://gist.github.com/871167
18:30 andrei_ the else statement seems to be executed no matter what
18:32 andrei_ the @organism variable has the right value when the if is true, but then it changes when it gets to else
18:32 andrei_ when it should just jump over the else statement
18:32 andrei_ I don't understand why this happens...
18:34 pyrimidine andrei_: do you know that the first conditional is true? eg: if ($feat_object -> has_tag('organism'))
18:35 andrei_ yes
18:35 andrei_ because it is a genbank file and it has that tag
18:36 sheenams left #bioperl
18:36 andrei_ I have another genbank file that does not have that tag and therefore only applies to the else block
18:37 andrei_ in this case the code runs perfectly
18:37 andrei_ but in the first case, after the if is executed, else is executed, which changes the value of the variable from the organism name to some other code
18:39 pyrimidine andrei_: Do you have an example of both GenBank files?
18:40 andrei_ yes do you want me to post it?
18:40 pyrimidine you can use gist if they're not terribly large
18:43 andrei_ this is the one that should work with if https://gist.github.com/871214
18:44 andrei_ https://gist.github.com/871211 and this is the else
18:44 andrei_ as you can see, in the first one, the organism name is under organism, and in the second one it is under accession
18:45 pyrimidine ah, I see the problem, you are mixing up annotation and seqfeatures
18:45 dbolser sheenams!
18:45 dbolser missed him
18:46 pyrimidine andrei_: http://www.bioperl.org/wik​i/HOWTO:Feature-Annotation
18:47 pyrimidine The organism information currently isn't stored as a feature, but in a special class (Bio::Species)
18:47 pyrimidine andrei_: http://www.bioperl.org/wiki/HOWTO:Fe​ature-Annotation#The_Species_Object
18:49 andrei_ but how can I get the accession which is in annotation, from a seq feature?
18:51 andrei_ oh I think i got it its not the accession that is the problem, but the Organism
18:51 pyrimidine andrei_: yep
18:51 pyrimidine andrei_: the accession and the species object are both stored in the $seq_object
18:51 pyrimidine $seq_object->accession_number
18:51 pyrimidine and
18:51 andrei_ yes,
18:52 pyrimidine $seq_object->species
18:52 andrei_ ok
18:52 andrei_ but why would it temper with the execution of if...else?
18:52 pyrimidine the first block never executes
18:53 pyrimidine (there is no seqfeature tagged 'organism')
18:54 andrei_ I did a print @organism; at both blocks and in the first it output the organism and when it got to the second print@organism from else it changed it to another value...This is what messed me up
18:55 pyrimidine Well, @organism isn't localized to the loop
18:55 pyrimidine is it possible the values are carried over from previous iterations?
18:56 andrei_ oh yes thats true there is more to the code, but still, that is the first time @organism is called
18:58 pyrimidine try localizing it to see if it changes anything
19:01 * pyrimidine -> lunch
19:11 dbolser has anyone ever used all 885 bioperl modules in one script?
19:11 deafferret yee gods  :)
19:11 dbolser ;-)
19:11 dbolser I'm working on a deobfusticator of sorts
19:12 dbolser /homes/dbolser/perl5/lib/p​erl5/Bio/Nexml/Factory.pm <- errors out
19:15 pyrimidine dbolser: POD issues?
19:15 dbolser pyrimidine: always
19:15 dbolser how do I | grep 2> again?
19:17 dbolser BEGIN failed--compilation aborted at /homes/dbolser/perl5/lib/p​erl5/Bio/Root/Test/Warn.pm line 82.
19:18 dbolser ?
19:22 dbolser managed to use all but 25
19:23 pyrimidine Bio::Root::Test is just a convenience class, but I think it has some problems.
19:23 dnewkirk joined #bioperl
19:23 dbolser pyrimidine: here is the list of BPMs that failed to 'use' on my box:
19:24 dbolser http://pastebin.com/BL0XYsJP
19:24 dbolser mostly usual suspects
19:24 dbolser bowtie is installed, but it doesn't find it
19:25 pyrimidine I think it's mainly loading issues; the bowtie module I think inherits from Bio::Assembly::IO::sam
19:26 dbolser ic
19:26 pyrimidine the others probably have similar issues
19:28 dbolser bleah, didn't realise some modules are dieing
19:29 dbolser http://pastebin.com/eYg4DTKt
19:29 andrei_ ok I modified the code and it seems to work
19:29 dbolser do I just need to grep for those messages to find the offending modules?
19:29 dbolser Bio::Ext::Align I guess
19:30 andrei_ the only problem now is that if the file has no species info, the script just dies...
19:32 pyrimidine dbolser: The Bio::Tools::EUtilities one is odd
19:33 * pyrimidine -> $job
19:33 dbolser I just commented out the last 400 packages and it starts 'working'
19:33 dbolser gl
19:35 dbolser fails at the first introspection
19:35 dbolser my $meta = Class::MOP::Class->initialize(package => $class); # $class = Bio::Align::Utilities
19:36 dbolser woot!
19:42 dbolser Bio::Variation::AAChange ... Can't use string ("-1") as a symbol ref while "strict refs" in use at /homes/dbolser/perl5/lib/perl5/i386-linux-​thread-multi/Class/MOP/Mixin/HasMethods.pm line 207, <BPM> line 834.
19:45 dbolser sweepstake... how many methods in bioperl?
19:45 dbolser no peaking! guesses into me...
19:47 dbolser https://github.com/dbolser/biomop
19:47 dbolser night
19:48 * dbolser wonders why README isn't updated
19:49 dbolser how do I ... meh, I'll just commit again
20:14 perl_splut joined #bioperl
20:50 50UAAC3D9 joined #bioperl
21:04 reecehart left #bioperl
21:14 50UAAC3D9 left #bioperl
21:19 CIA-57 bioperl-live: Chris Fields topic/cached_locations * rd4e270e / (Bio/Factory/FTLocationFactory.pm Bio/PrimarySeqI.pm): add a bit of caching magic; with large sequence parses we get about 20-50% faster, with the downside we will ened to reset the cache between sequence parses - http://bit.ly/gMLBSA
21:19 pyrimidine *need
21:20 cassj joined #bioperl
21:26 * dnewkirk wonders what pyrimidine needs
21:26 pyrimidine dnewkirk: just a misspelling in my commit msg, that's all
21:27 * pyrimidine -> home
23:29 reecehart joined #bioperl
23:33 dnewkirk left #bioperl
23:34 reecehart left #bioperl

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