Camelia, the Perl 6 bug

IRC log for #bioperl, 2010-05-06

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

All times shown according to UTC.

Time Nick Message
00:01 deafferret (i get same result as you)
00:07 abooz mm..When i tried the code with the test.gff file from bioperl-1.2.3/t/data/biodbgff/ i got even more errors.
00:12 deafferret ah. is your test.gff tab-delimited?
00:14 abooz lemme check.
00:14 deafferret this works for me   http://github.com/jhannah/​sandbox/tree/master/abooz/
00:14 * deafferret hates tabs with a passion
00:28 abooz hmm..getting the same error.
00:29 abooz wondering if my sed didnt work..
00:30 deafferret GFF is tab only  :(
00:32 abooz Ah, ok works now.No stupid message.
00:32 abooz Thanks!
00:32 deafferret :)
00:35 abooz one more question :)..why do i only get the seq_id? No source ,type, i only get the object ref for them
00:36 luisrodr joined #bioperl
00:38 deafferret same code?
00:41 abooz yep.
00:43 abooz i hate objects in perl....
00:44 deafferret :)
00:44 deafferret seq_id is 'Chr1' ?
00:44 abooz it's the first column so it should be.(i think!)
00:46 deafferret $feature->get_tag_values('source')
00:47 deafferret http://www.bioperl.org/wik​i/HOWTO:Feature-Annotation
00:47 deafferret $feature->get_tag_values('type')
00:50 abooz hmmm....why is seq_id() working then?
00:51 deafferret dunno...
00:56 deafferret your .gff had errors. I fixed it. And here's a full parse run:    http://github.com/jhannah/​sandbox/tree/master/abooz/
01:06 brandi joined #bioperl
01:06 brandi left #bioperl
01:10 abooz is it possible to see a diff on github?
01:12 abooz got it.
01:14 abooz cheers,Don't know how there errors creeped in there.
01:14 deafferret solar flares, probably  :)
03:06 aendruk joined #bioperl
03:34 deafferret heehee  http://jays.net/wiki/Smolder
05:59 rbuels oh boy, deafferret is now the official bio smoke tester.
06:01 rbuels deafferret: how often will you smoke?
06:01 rbuels deafferret++ # by the way
07:36 luisrodr joined #bioperl
08:42 bag joined #bioperl
10:20 driveby_bot joined #bioperl
10:20 driveby_bot /home/svn-repositories/bioperl: r16963 (dave_messina) : prevent PAML parser from prematurely processing seqs. (provide possibility for properly parsing pre-PHYLIP header lines like "Before deleting alignment gaps").
10:20 driveby_bot Diff: http://tinyurl.com/36gk4uy
11:32 brandi joined #bioperl
11:33 brandi left #bioperl
12:27 bag joined #bioperl
13:14 brandi joined #bioperl
13:20 brandi left #bioperl
13:37 cawiss left #bioperl
13:38 cawiss joined #bioperl
15:08 spekki01 joined #bioperl
15:11 bag joined #bioperl
15:38 deafferret rbuels: :) next step is figuring out how to add it as an install option to Build.PL i guess. my build foo is weak
15:38 deafferret my testing foo is getting pretty good
15:38 deafferret congrats on Core Dev! woot!
15:38 deafferret this IRC channel is all respectable or something now
15:39 deafferret & motorcycle pickup
18:10 melic joined #bioperl
18:46 abooz I've got more questions about gff parsing. I'm getting an error while using next_feature_group. I'm getting EXCEPTION: Bio::Root::Exception MSG: Parent with ID 13101.m00002 not found.
18:46 abooz I'm wondering if the way the second isoform of the protein is in the file has problems...
18:47 rbuels abooz: probably the file is not conforming to the gff3 spec in some way
18:51 abooz can i have a parent relationship with a feature that comes after a ###?
19:02 rbuels abooz: the ### divider explicitly tells the parser that nothing after the ### can have a parent relationship with anything above the ###
19:03 rbuels this is an optimization to limit memory usage.
19:03 rbuels if your gff3 is small, you could remove the ### dividers
19:04 rbuels otherwise, you'll probably need to fix them
19:05 abooz nope, it's quite large.
19:23 bag joined #bioperl
19:26 abooz why each different institute has its own way of implementing the same format?
19:26 abooz i'm getting fed up with it.
19:27 perl_splut because they didn't talk with each other as they developed them in house
19:28 abooz well, when you implement somebody else's format then why not conform to it?
19:30 perl_splut depends on how well they were versed in the format
19:31 perl_splut one "developer" in my lab kept trying to add data willy-nilly to a FASTA header line and couldn't understand why the formatdb command wasn't using his new information in the DB
19:37 rbuels abooz: the reason for all the fragmentation is that there is no incentive to cooperate.  it is not rewarded by funding agencies.
19:37 perl_splut that too
19:38 rbuels also there is no incentive to polish and debug and maintain software, for the same reason.
19:38 perl_splut that and grad students finish what they were doing and move on
19:39 rbuels and they have no incentive to maintain what they left behind, because there is no reward in that.
19:40 perl_splut yep. I've done that with quite a bit of stuff I've written
19:40 perl_splut got it working. got the data. on to other things
19:41 perl_splut funny when someone comes up to me a few years later and asks me questions about how the code works.
19:41 perl_splut in most cases I've completely forgotten writing it, let alone how it does what it does
19:44 rbuels should the academic system put more emphasis on software?  i'm not sure.  possibly.
19:44 perl_splut depends on the degree the writer was going for
19:45 perl_splut in my lab there are only 2 people that I would even consider computer literate
19:46 perl_splut the rest can barely get excel to work and do so many things the hard way
19:47 rbuels i think 2 semesters of CS should be part of the Bio core curriculum
19:47 rbuels or one semester, at least.
19:48 rbuels for the amount that biologists use computers, they really should have some clue about them.
19:48 rbuels computers are critical nowadays for every. part. of. biology.
19:48 perl_splut the only ones that have that forced on them are those who started out going for their undergrad degree at this university
19:49 perl_splut and that is only CSE100 (computers for the really dumb)
19:49 rbuels your university has that requirement?  good on it!
19:49 rbuels which university?
19:50 perl_splut by the end of the semester you should know how to change fonts in word. Change cell colors and a few, VERY BASIC, functions in excel and make "pretty" pictures in Powerpoint
19:50 perl_splut ASU
19:50 perl_splut but most of the people I work with didn't do their undergrad here
19:50 * rbuels nods
19:50 perl_splut in fact, they are at a Masters or above
19:51 perl_splut with most being technicians and research scientists, not grad, post-grad or undergrad workers
20:39 faceface joined #bioperl
20:41 deafferret ..o.
20:44 deafferret ha, i love a good rant against academia. that's how my programmer brain always reads those critiques, anyway
20:53 perl_splut In a world where productivity doesn't matter, the minutiae become far too important
20:54 deafferret lol
20:57 deafferret anyone wanna come dirt biking with me this weekend? my usual victims are all unavailable
20:57 deafferret flights to OMA are cheap!
20:58 perl_splut oma?
20:59 deafferret Omaha, NE, USA. Center of Everything!
20:59 perl_splut and the middle of nowhere
21:00 deafferret tsk tsk
21:00 perl_splut :)
21:00 deafferret the OMA chamber of commerce is going to write you some nasty faxes now
21:02 perl_splut heheh
21:02 faceface joined #bioperl
21:36 melic joined #bioperl
21:54 pyrimidine joined #bioperl
22:03 melic joined #bioperl
22:12 Biome joined #bioperl
22:12 Biome biome: cleanup_roles Chris Fields * b6dd1ff (1 files in 1 dirs): test post-commit github hook - http://bit.ly/94S6gz
22:12 Biome left #bioperl
22:34 dnewkirk_hl joined #bioperl

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