Camelia, the Perl 6 bug

IRC log for #bioperl, 2011-03-31

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

All times shown according to UTC.

Time Nick Message
00:14 gylz hi
00:14 gylz can somebody help me with a code?
00:20 jhannah probably. :) did you read the topic?
00:23 gylz no i didn't
00:23 gylz why?
00:23 gylz Anyway......
00:23 gylz I'm trying to make a program
00:24 gylz that extracts the cds sequences
00:24 gylz from a set of databank files
00:26 gylz This is the code I used for making against the NCBI
00:26 gylz #! /usr/bin/perl use strict;  use Bio::DB::Query::GenBank; use Bio::DB::GenBank;        my $query_string = '(acetolactate synthase[All Fields] AND "green plants"[porgn]) AND "flowering plants"[porgn] AND "complete cds"[All Fields]';    my $query = Bio::DB::Query::GenBank->new(-db => 'nucleotide', -query => $query_string);     print "Number of entries: ", $query->count,"\n";     # get a Genbank database handle    my $gb = Bio::DB::GenBan
00:27 gylz the while loop is...............
00:27 gylz while (my $seq = $stream->next_seq) {       print "Sequence ",$seq->id, " first 10 bases ", $seq->subseq(1,10), "\n";    }
00:28 gylz however it throws an exception
00:29 gylz Something like this...
00:29 gylz MSG: Unexpected error in feature table for  Skipping feature, attempting to recover
00:29 gylz I have google it already...
00:30 gylz and it isn't new...
00:31 gylz but i don't know how to fix it...
00:32 gylz (for example... http://ugene.unipro.ru/foru​m/YaBB.pl?num=1253431935/4)
00:40 gylz The code is indeed from http://doc.bioperl.org/releases/bioperl-cur​rent/bioperl-live/Bio/DB/Query/GenBank.html
00:41 gylz changing the loop block...
00:54 jhannah dont paste code into channel. use gist.github.com
00:54 jhannah per the topic
00:54 jhannah that you read  :)
00:55 gylz sorry, thanks for the point
00:58 gylz The code... in gist
00:58 gylz git://gist.github.com/895625.git
00:59 jhannah much better. thanks  :)
00:59 gylz Thank you again...
00:59 jhannah you can also use that same gitst to list the error(s) at the bottom (tip for next time)
01:00 jhannah what version of bioperl are you running?
01:00 gylz i think 1.6
01:00 gylz let me see
01:00 gylz I'm mostly from the java world...
01:00 jhannah put the entire output at the bottom of your gist
01:01 jhannah "MSG: Unexpected error " is not very helpful, i'm hoping there's more specific error output in there somewhere  :)
01:02 gylz I've appended the output...
01:03 jhannah excellent!  :)
01:05 jhannah does it keep going after that MSG? or die?
01:05 gylz die
01:06 jhannah what happens if you shorten the print line to just   print "Sequence ",$seq->id, "\n";    ?
01:06 gylz let me try...
01:06 jhannah i'd also try directing STDERR to STDOUT like     perl myprogram.pl 2>&1
01:07 jhannah to (maybe) get the error to show up in the true location
01:07 gylz indeed the original while have many more sentences but i reduced them to that one...
01:08 jhannah sure. i'm just trying to isolate if its ->subseq that's failing
01:10 gylz I'm running it without the  " first 10 bases ", $seq->subseq(1,10), "\n";
01:11 jhannah and?
01:11 gylz It's still running...
01:12 gylz some of the sequences are quite large...
01:12 * jhannah nods
01:15 gylz Lot of HD scratching
01:15 gylz And finally die again...
01:17 gylz https://gist.github.com/895650
01:19 gylz may I have to put the while into a eval or a try block...
01:19 gylz ????
01:27 jhannah gak. bioinformatics data sucks  :/
01:28 jhannah i wonder if the AP in AP008208 is significant and should be avoided.
01:28 jhannah you might want to email https://gist.github.com/895650 to the bioperl-l mailing list
01:29 gylz ok
01:29 gylz thanks you again
01:29 jhannah ya, sorry i didn't know. I tend not to  :)
01:30 gylz By the way...
01:31 gylz the exception
01:31 gylz management system
01:31 gylz works like in java...
01:31 gylz it evals a block...
01:31 gylz some function, class or whatever
01:32 gylz throws an exception
01:32 gylz and a catch (or the if($@))
01:32 jhannah well are you wanting to skip it or actually get to the bottom of the problem?
01:32 gylz take it...
01:32 gylz my question it
01:32 gylz after catching it
01:32 gylz dies...
01:33 gylz or continues the normal flow...
01:33 jhannah this one is popular nowadays   http://search.cpan.org/perldoc?Try::Tiny
01:33 gylz oh...
01:33 gylz thanks
01:34 jhannah or, old school built in is   perldoc -f eval
01:39 rdesfo left #bioperl
01:50 gylz left #bioperl
02:46 mmlevitt joined #bioperl
04:26 mmlevitt left #bioperl
04:37 mmlevitt joined #bioperl
04:51 zenman_ joined #bioperl
05:00 zenman_ left #bioperl
06:21 mmlevitt left #bioperl
06:26 mmlevitt joined #bioperl
06:58 mmlevitt left #bioperl
07:06 mmlevitt joined #bioperl
08:44 perl_splut left #bioperl
08:49 perl_splut joined #bioperl
08:52 cassj joined #bioperl
11:50 mmlevitt left #bioperl
12:23 cassj left #bioperl
13:02 cassj joined #bioperl
13:02 cassj left #bioperl
13:28 zenman joined #bioperl
13:36 zenman left #bioperl
14:31 rdesfo joined #bioperl
14:38 dgk joined #bioperl
14:40 dgk Hello! I need to parse a psiblast report, but the BPpsilite module is deprecated, it seems. When parsing like a normal BLAST, I get repeated results, because of the iterations. How can I have my code show only the results from the last iteration?
14:43 pyrimidine dgk: have you looked at Bio::SearchIO::blast?
14:43 pyrimidine It handles iterations I believe
14:44 dgk Yes, but I couldn't find anything on iterations. Whatever I could find was with the BPpsilite module. I'll take a look again, then.
14:45 pyrimidine dgk: if you have the full code checkout, look at the test suite, specifically t/SearchIO/blast.t
14:45 dgk Ok, I will. Thank you very much!
14:45 pyrimidine dgk: and look for 'next_iteration'
14:46 dgk Ok, thank you.
15:02 CIA-119 bioperl6: takadonet master * r58ee396 / (3 files in 2 dirs): Fixed a bug in LocationFactory for region with multiple join,order or bond in $sub section. Implemented a few more simple function in Lite.pm with corresponding tests - http://bit.ly/gCwvqg
15:07 dgk left #bioperl
15:47 dbolser pyrimidine: is it too late for me to object to the way bioperl tutorials are maintained on the wiki and in source code?
15:48 pyrimidine dbolser: there are definite shortcomings to having the tutorials on the wiki
15:48 pyrimidine but what are your thoughts?
16:06 dbolser yeah, its borderline, so I don't want to make a big deal
16:09 dbolser but the 'flagged revs' mediawiki extension is very good here. People who are not logged in see the latest 'official' version of a particular page by default. People who are logged in see the bleeding edge version of a particular page by default. Special 'dev' and 'doc' users have the final review on all contributors edits.
16:10 dbolser so questions and bad examples added to the docs don't show up for most. They are added to teh list of edits to review for the special users, and everyone can contribute.
16:11 dbolser for example, as a user with the 'review' permission, if I edit a page, I 'auto accept' all changes since the last 'flagged rev'.
16:11 dbolser the module build script would bring in only the latest flagged rev when adding wiki to the PM
16:12 dbolser module builders would be required to create tutorials on the wiki... yah... but it screws up pod for the module itself I guess
16:12 dbolser I just figure, more authors + good management = better docs
16:13 dbolser more authors + no or bad management = terrible docs, so it has to be committed to
16:13 dbolser otherwise, keep docs off the wiki
16:13 dbolser I don't see any advantage of just cloning them there
16:14 dbolser ok, my belated $0.02
16:16 pyrimidine dbolser: I think the consensus of the devs is to start moving the docs back into the distribution at some point
16:16 pyrimidine somewhat like Bio::Manual::*
16:17 pyrimidine which may come with the added benefit of allowing inline testing
16:18 pyrimidine The wiki would likely have an autoupdated version of those
16:58 CIA-119 bioperl6: takadonet master * ra08e1c9 / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t):
16:58 CIA-119 bioperl6: New attribute: ref,phase,class,subtype and array segments. Now can handle
16:58 CIA-119 bioperl6: segments in new constructor. Implemented add_segments,segments,seq_id,phase and
16:58 CIA-119 bioperl6: class methods. Most will probably go away since they are simple getter/setter
16:58 CIA-119 bioperl6: methods. Added more tests. - http://bit.ly/gW7lZE
16:58 rdesfo left #bioperl
17:02 CIA-119 bioperl6: takadonet master * ra972417 / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t): Implemented and tested for feature_count - http://bit.ly/feluqi
17:17 takadonet joined #bioperl
17:17 takadonet hey all
17:22 * rbuels waves
17:23 pyrimidine takadonet: busy? :)
17:23 takadonet pyrimidine: nope, not working today
17:23 takadonet just hacking on bioperl6
17:24 CIA-119 bioperl6: takadonet master * rd608201 / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t): Implemented and tested is_circular attribute/methods - http://bit.ly/gfCbfx
17:36 * jhannah lurks unseen
17:38 CIA-119 bioperl6: takadonet master * rb4fbfcd / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t): Testing parts of segments method that accepts other Lite objects instead of array reference of start/stop coordinates. Also ensure that they keep their own type - http://bit.ly/ezI37F
18:02 CIA-119 bioperl6: takadonet master * rff18ffb / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t): Implemented and added test for method each_Location. - http://bit.ly/f2u28L
18:16 pyrimidine jhannah: what happened to deafferret?
18:22 * rbuels thinks jhannah probably ate him
18:22 rbuels poor little ferret.
18:22 * rbuels sniffs
18:23 * jhannah belches
18:23 jhannah forgot to stick that alias in my new irssi config
18:24 jhannah is now known as deafferret
18:26 deafferret https://github.com/jhannah​/sandbox/tree/master/irssi
18:27 CIA-119 bioperl6: takadonet master * r17c7e2d / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t): Implemented low, high, min_start,max_start,min_end,max_start and to_FTstring. Added tests for low,high and to_FTstring - http://bit.ly/gaTBbf
18:54 CIA-119 bioperl-live: Chris Fields master * r992529f / README : test post-receive again - http://bit.ly/hglYPq
18:55 CIA-119 bioperl-live: DaveMessina master * rcdf38e4 / README : Merge branch 'master' of github.com:bioperl/bioperl-live - http://bit.ly/eXHv4G
19:07 CIA-119 bioperl6: takadonet master * r54c94f4 / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t): format fixing and changed # of tests - http://bit.ly/euz8mc
19:07 CIA-119 bioperl6: takadonet master * r6fb6f36 / (3 files in 3 dirs): Added an extra ref check in Split.pm. Just a hack for now. Implemented Location and end methods. 5 more tests were created. - http://bit.ly/eN9Bli
19:10 dave_messina joined #bioperl
19:11 dave_messina Awright, could someone please tell me how to commit without fucking up the history? :)
19:11 takadonet which project?
19:12 dave_messina bioperl-live
19:12 * takadonet points to pyrimidine
19:14 pyrimidine huh?
19:15 * pyrimidine sees no history fuck-upery
19:15 pyrimidine dave_messina: everything ok?
19:15 dave_messina my last commit...I managed to 'merge branch master'
19:15 rbuels dave_messina: you'll probably want to grok the difference between git pull and git pull --rebase
19:15 rbuels dave_messina: but the merge branch master thing isn't a big deal
19:15 pyrimidine what rbuels said
19:15 dave_messina oh really? okay. cool.
19:15 dave_messina and I'll check out pull --rebase.
19:16 dave_messina thanks guys
19:16 pyrimidine FYI, I'm reworking the redmine repo, may see some slowdown on that URL
19:16 dave_messina ok.
19:21 CIA-119 bioperl6: takadonet master * rbb83cd1 / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t): Implemented location_string method and added 3 tests - http://bit.ly/esvAvk
19:21 CIA-119 bioperl6: takadonet master * r45a1062 / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t): Score attribute with corresponding method are done. Deleted p5 code of coordinate_policy. Added a few more tests - http://bit.ly/fqzSv4
19:33 CIA-119 bioperl6: takadonet master * rd40f3c1 / (lib/Bio/SeqFeature/Lite.pm t/SeqFeature/Lite.t): Implemented length method and added tests for them - http://bit.ly/edy7mx
19:35 dave_messina left #bioperl
19:47 CIA-119 bioperl-live: Chris Fields master * rb3c6342 / README : test hook again - http://bit.ly/ewqhav
20:05 micrypt joined #bioperl
20:07 dave_messina joined #bioperl
20:08 CIA-119 bioperl-live: Chris Fields master * rd5de022 / README : test hook again - http://bit.ly/hxn7O6
20:10 mmlevitt joined #bioperl
20:16 takadonet left #bioperl
20:25 dave_messina left #bioperl
20:27 sheenams joined #bioperl
20:58 pyrimidine left #bioperl
20:59 micrypt HI, what does this mean? "WARNING- MSG: No HSPs for this minimal Hit...."?
21:07 deafferret what comes after ....?
21:10 micrypt deafferret: A series of errors similar to this: http://dpaste.com/527520/
21:12 deafferret what line of bioperl code generates that?
21:12 deafferret is your bioperl current?
21:12 * deafferret doubts he will be very helpful
21:13 micrypt deafferret: I could paste the whole file if you have the time.
21:14 deafferret i have no time, but pasting everything is always a good idea. someone might have the time here or on the bioperl-l mailing list
21:14 deafferret you don't know until you paste and ask  :)
21:16 micrypt deafferret: http://perl.nopaste.dk/p3081
21:17 deafferret for best results paste (1) your code (2) your errors (3) relevant input (if any) all in one paste
21:17 micrypt It gets a series of results from line 69 ($y_axis[$count] = $hit->score;) and fails for the last few.
21:17 micrypt Okay, just a sec'.
21:21 micrypt deafferret: Updated > http://perl.nopaste.dk/p3082
21:22 deafferret ah! an EXCELLENT inquiry!
21:22 deafferret If you send that to bioperl-l you'll probably get quick feedback
21:22 deafferret (i don't know)
21:23 deafferret micrypt++   :)
21:23 micrypt deafferret: The mailing list?
21:23 deafferret ya
21:23 micrypt Okay, thanks for checking it out.
21:24 deafferret ya, providing all the info for the question is usually time consuming on the mailing list so hopefully i saved you a bunch of time  :)
21:51 sheenams left #bioperl
22:07 dgk joined #bioperl
22:08 dgk Hello! Is there a way when doing a standalone blast to only show results from the same genus?
22:10 mmlevitt left #bioperl
22:10 mmlevitt joined #bioperl
22:21 dgk left #bioperl
22:46 micrypt left #bioperl
23:17 micrypt joined #bioperl

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