Camelia, the Perl 6 bug

IRC log for #bioperl, 2010-06-07

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

All times shown according to UTC.

Time Nick Message
00:46 brandi joined #bioperl
00:47 brandi left #bioperl
03:19 CIA-94 bioperl-live: Chris Fields master * r4b3399d / (Bio/Location/Atomic.pm Bio/Location/Simple.pm): add is_remote docs from Bio::LocationI to Atomic and Simple - http://bit.ly/af5iuv
03:20 CIA-94 bioperl-live: Chris Fields master * rbcb335e / (Bio/Location/Simple.pm Bio/LocationI.pm): small tweak to add for Args in docs - http://bit.ly/bVVCyq
03:44 CIA-94 bioperl-live: Chris Fields master * rde6316c / Bio/Search/Hit/GenericHit.pm : [bug 3089] iteration and found_again added to GenericHit for consistency (not applicable to anything but PSI-BLAST - http://bit.ly/9u0YEq
03:44 CIA-94 bioperl-live: Chris Fields master * r4561de1 / Bio/Search/Hit/GenericHit.pm : Merge branch 'topic/bug3089' - http://bit.ly/9nohZw
03:49 luisrodr joined #bioperl
04:05 CIA-94 bioperl-live: Chris Fields master * r368a23b / Bio/TreeIO/nexus.pm : [bug 3028] some rearrangement for FigTree output - http://bit.ly/9jRMTq
05:04 bag_ joined #bioperl
05:28 XaeroOne joined #bioperl
05:29 XaeroOne joined #bioperl
06:30 kai joined #bioperl
12:18 jyin joined #bioperl
12:19 jyin Hi, what is the best IRC software in Mac, I am using Colloguy, but it seems it is not as good as mIRC...
12:56 jyin It seems bioperl channel is not very active today, is it bank holiday too in the US?
13:05 * kai shrugs
13:05 kai some people actually work ;)
13:06 kai that said, I'm not a mac user, but I seem to remember using XChat aqua when I last irced on a mac
13:06 kai that's been years, though, before they switched to intel :)
13:08 brandi joined #bioperl
13:08 brandi left #bioperl
13:11 jyin ok, i see
13:13 jyin I think Colloguy is ok, it is just it doesnot look so professional as mIRC
13:14 Jun joined #bioperl
13:30 jyin got another question. For example, in git, I have two branches, master and testbranch.
13:31 jyin I switched to testbranch, created two files. Then I wrote "git add ." and "git commit -m "add file"".
13:31 jyin After that, i switch back to master "git checkout master"
13:33 jyin probably I added two new lines before I switched to testbranch
13:34 jyin Now I cannot switch back to testbranch.... It says "error: You have local changes to "xxxx.txt", cannot switch branch.
13:35 jyin I tried "git commit", but it seems it doesnot work ...
13:35 kai so what does "git diff" say these changes are?
13:36 kai in other words, do you care about those changes or are you happy to throw them away?
13:41 jyin Actually  I want to merge the testbranch with master
13:42 jyin "git diff" says quite a few things...
13:43 Jun --- a/test.txt
13:43 Jun +++ b/test.txt
13:43 Jun @@ -1,2 +1 @@
13:43 Jun -xxx
13:43 Jun -yyy
13:43 Jun +xxx
13:46 Jun And, if I use "git merge", it says "error: Your local changes to 'test.txt' would be overwritten by merge.  Aborting
13:46 Jun ."
13:47 kai well, you can't merge if you have local conflicting changes
13:47 kai uncommitted, that is
13:48 kai so you could git add test.txt
13:48 kai git commit -m "local changes in master"
13:48 kai and then merge
13:49 jyin Ok, thx. I think I need to read more on git documentation ... Git is much more complicated than I thought
13:50 Jun It works! Cheers
13:50 kai the learning curve is a bit steep, but I wouldn't want to switch back :)
13:51 jyin ok, it seem you should commit all changes before you switch ...
13:52 kai usually, yes
13:52 kai or stash them
13:52 jyin yep, I know. Git is a must-do ... I am sure I will enjoy it after I master it :D
13:52 kai have a look at git stash
13:52 kai that said, I don't like git stash much, but I know other people do
13:53 jyin ok~
13:54 kai I'm more of a "git diff > ../feature.patch; git apply -R < ../feature.patch; git checkout other_branch" person
13:55 kai o, time to head home, bbl
14:07 melic joined #bioperl
14:25 melic joined #bioperl
14:32 jyin bye
14:34 melic joined #bioperl
14:47 jyin left #bioperl
17:46 deafferret .o.oooo.O
17:48 rbuels it's a bit bubbly.
18:21 deafferret the poop bubble tickle my nose!  -giggle-
19:05 CIA-94 bioperl-live: Robert Buels topic/longest_orf * r1ef6c24 / Bio/Tools/CodonTable.pm : added comment (not pod) block for internal _codon_is in CodonTable.pm - http://bit.ly/blMGD0
19:05 CIA-94 bioperl-live: Robert Buels topic/longest_orf * r9e96c95 / Bio/PrimarySeqI.pm : clarified some translate() POD - http://bit.ly/bxxnV7
19:05 CIA-94 bioperl-live: Robert Buels topic/longest_orf * r6e42163 / Bio/PrimarySeqI.pm : PrimarySeqI.pm whitespace cleanup - http://bit.ly/cAR0lw
19:09 kblin evening folks
19:11 pyrimidine joined #bioperl
19:13 CIA-94 bioperl-live: Robert Buels master * r1ef6c24 / Bio/Tools/CodonTable.pm : added comment (not pod) block for internal _codon_is in CodonTable.pm - http://bit.ly/blMGD0
19:13 CIA-94 bioperl-live: Robert Buels master * r9e96c95 / Bio/PrimarySeqI.pm : clarified some translate() POD - http://bit.ly/bxxnV7
19:13 CIA-94 bioperl-live: Robert Buels master * r6e42163 / Bio/PrimarySeqI.pm : PrimarySeqI.pm whitespace cleanup - http://bit.ly/cAR0lw
19:13 CIA-94 bioperl-live: Robert Buels master * rfa43e16 / (3 files in 3 dirs):
19:13 CIA-94 bioperl-live: Merge branch 'topic/longest_orf' into moster
19:13 CIA-94 bioperl-live: Conflicts:
19:13 CIA-94 bioperl-live:  t/Seq/PrimarySeq.t - http://bit.ly/9Oye0y
19:13 pyrimidine I hate it when I merge in moster...
19:14 pyrimidine Kinda hurts
19:14 rbuels pyrimidine: lol.  that was my new-upstream-master
19:14 rbuels pyrimidine: forgot that was in the commit message
19:14 rbuels somehow my main local master got screwed up so i made a new one for the merge and called it moster
19:15 pyrimidine rbuels: you are a real peng in the neck
19:15 pyrimidine see that?  that was a pun.
19:15 rbuels oooh, oh yeah.
19:15 rbuels i get it.
19:15 rbuels right.
19:16 pyrimidine also laced with sarcasm, a side of annoyance, and a dash of mst
19:16 deafferret merge branch: conflicts?   moster conflicts! eek!
19:17 rbuels i should have called it monster.
19:17 pyrimidine yes
19:17 pyrimidine exactly
19:17 deafferret monster mash. it was a graveyard smash!
19:17 deafferret merge master? no. mash monster.
19:18 deafferret ATTENTION: The women's restroom is closed. That is all.
19:18 deafferret email++
19:19 CIA-94 bioperl-live: Robert Buels master * r75ab713 / (20 files in 13 dirs): Merge branch 'master' of github.com:bioperl/bioperl-live - http://bit.ly/9j4rdS
19:20 rbuels eh?  there was supposed to be a whitespace commit in there.
19:20 * rbuels thinks the github->CIA integration is not always totally correct
19:20 pyrimidine prob. the cia bot
19:21 pyrimidine rbuels: if you have a CIA account, the bot is public.  It can be customized a bit.
19:21 pyrimidine http://cia.vc/account/bots/17973/
19:24 kblin hey pyrimidine
19:24 pyrimidine o/
19:25 kblin is there any chance to get the hmmer3 parser included in that fabled 1.6.2 release about which I heard murmurs on the mailing list?
19:26 pyrimidine depends
19:26 pyrimidine does it introduce any dependencies?
19:26 pyrimidine how are the tests?
19:26 pyrimidine etc etc etc
19:28 pyrimidine rbuels: can't remember, are you going to BOSC?
19:29 rbuels pyrimidine: not planning on it.  i'm already running around like a chicken with my head cut off.  (to use a favorite Indiana simile)
19:29 pyrimidine ok
19:29 * rbuels actually does not know whether that simile is favored by the residents of the great state of Indiana
19:29 kblin pyrimidine: dependencies: not that I'm aware of, it's pretty similar to what the hmmer2 parser does
19:30 pyrimidine rbuels: they say the same thing in texas
19:30 rbuels which is where i learned it  :-)
19:30 pyrimidine kblin: then the tests become critical
19:30 kblin pyrimidine: it probably could do with more tests, but if that's the criterium of getting it into the release, I can certainly make some time for that
19:30 pyrimidine rbuels: that 'learnt' in texas
19:30 rbuels yeah i was about to correct my spelling
19:31 rbuels pyrimidine++ # quick draw
19:31 pyrimidine kblin: basically, it would need to parse the widest range of output possible
19:32 pyrimidine any variations expected
19:32 pyrimidine though that's hardly easy to take into consideration
19:32 kblin it parses all the files I've seen so far :)
19:32 pyrimidine good!
19:32 rbuels tests.
19:33 pyrimidine so, maybe test against any output variations.  that's all it really needs.
19:33 pyrimidine can look at test coverage as well
19:33 rbuels i would actually say that a hmmer3 output file definitely needs to be in the test suite
19:33 rbuels and the same tests should be run on the hmmer3 file as the hmmer2 files
19:34 pyrimidine rbuels: I'm thinking a few files, any variants we may run into, but keeping it to a minimum
19:34 pyrimidine but, yes, running the same tests if possible on both hmmer2 and hmmer3 output
19:34 rbuels a minimum in size and runtime, but not a minimum in coverage, lol
19:34 kblin pyrimidine: there's two commands that produce slightly different output, hmmscan and hmmsearch
19:35 pyrimidine 'overly-optimized' is my middle name
19:35 kblin I think I have example files in my repo
19:35 kblin hang on
19:35 * pyrimidine hanging on...
19:36 kblin hm, I was sure I had a hmmsearch test
19:37 kblin ah, that's the branch
19:38 kblin or not
19:38 kblin ok, I'm sure I've got it somewhere, just need to find it and push it to my github repo :)
19:38 * pyrimidine searches his pockets for it
19:39 * pyrimidine looks under the couch cushions
19:40 pyrimidine man, perl6 is coming along at a good clip.  I think the alpha will have quite a bit implemented
19:40 pyrimidine slow, but implemented
19:40 kblin but it's easy to redo if I don't find it
19:41 kblin the test I have cover pretty much what the hmmer2 tests cover as well, but I can make sure they get even closer
19:42 kblin I prefer to spend my time getting the parser into the distribution, rather than spending it on figuring out how to best distribute a patched version of bioperl to my users
19:43 pyrimidine we can work on gitting it in
19:43 pyrimidine see what I did there?  another pun.
19:43 rbuels oh, ho ho.
19:44 rbuels you're punchy today.
19:44 kblin sweet, I'll polish up the tests some more then
19:44 * pyrimidine feels like a boxer
19:44 kblin notice how, with me being german, I didn't notice the humour
19:45 rbuels because germans are humorless, yes.
19:45 rbuels i knew that.
19:46 kblin no, we just don't laugh without first getting a permit
19:46 rbuels lol
19:46 pyrimidine no, that's the british
19:46 kblin pyrimidine: they _queue_ to get one
19:46 * rbuels rolls his eyes
19:46 pyrimidine :)
19:46 rbuels and in america, if you don't get one, they send you to guantanamo.
19:47 kblin ah, vacation in cuba, great thing
19:47 * rbuels chomps the end off a cigar
19:47 kblin anyway, back to some useful work
19:47 kblin like searching for my MIA hmmsearch test
19:49 rbuels ah, it didn't get a permit.
19:49 rbuels you may have to search in cuba.
19:51 kblin nah, it was hanging out in another branch, with the cool kids
19:54 deafferret the cool kids use mercurial
19:54 kblin mercurial and I don't get along
19:55 deafferret <humor reader_lang_code="de">the cool kids use mercurial</humor>
19:55 kblin deafferret: it's ok, I don't like the cool kids either
19:56 deafferret :)
19:56 deafferret kblin: hows your hmmer3 takeover of bioperl coming?
19:57 kblin I think the test I just added found a bug. :)
19:57 kblin pretty good otherwise
19:58 deafferret kblin++
19:58 deafferret Jun: hows your "global domination via git" project coming?
19:59 pyrimidine what's the time there?  It's probably late.
20:00 pyrimidine 6-7 hrs difference, IIRC
20:00 pyrimidine oh, bioperl lost a facebook fan this week.
20:00 * pyrimidine sheds a tear
20:02 deafferret well, it was a good run. shut it down, boys.
20:02 * deafferret shoots BioPerl in the head
20:02 * pyrimidine plays taps
20:02 * deafferret went to a "practical shooters" event yesterday. pretty neat stuff.
20:05 * pyrimidine thinks of the movie 'trainspotting' when I hear the term 'practical shooters'
20:06 deafferret :)  http://www.youtube.com/watch?v=RoXNnqQWBKI
20:07 pyrimidine jesus, are they shooting some of those targets at point-blank range?
20:07 deafferret ...no?   i guess I'll have to actually watch it now
20:08 deafferret oh, crap   yes
20:08 deafferret never seen them shoot closer than ~15 ft before  :)
20:09 deafferret step 1: stick the gun in this dudes chest...   :)
20:09 pyrimidine about 53 secs in, he caps one in the gut
20:09 pyrimidine not even sopranos style
20:10 bag_ joined #bioperl
20:10 pyrimidine just quick draw, then blammo
20:10 deafferret accuracy should be pretty good at 0 feet
20:10 pyrimidine true ;>
20:10 deafferret i think you also lose points if you shoot yourself in the leg
20:11 deafferret pro tip: do not attempt to rob anyone at a gun club
20:12 pyrimidine or in a school for samurai
20:12 pyrimidine okay, btw
20:13 Rakmakallan joined #bioperl
20:15 kblin ok, looks like the hmmer3 parser does not like result files without alignments
20:15 kblin that doesn't look right
20:18 pyrimidine yes, it's all those little things that make developing a parser fun
20:18 * pyrimidine thinks wistfully back to when I developed the infernal parser
20:19 kblin hm, though it looks like the hmmer2 code is the same
20:19 Rakmakallan good evening. I'm making a perl script for using kyte-doolittle hydrophobicity tables in uniprot file to predict transmembrane sequences but i'm having some trouble
20:20 pyrimidine ...and the trouble is?
20:22 * deafferret rechecks his tequilaphilicity tables
20:22 Rakmakallan well there must be sth wrong in the loops because the numerical results i get through the script don't match the manual ones
20:22 Rakmakallan however, i can't seem to pinpoint it
20:29 kblin can you paste your code to codepad? (see link in topic)
20:29 kblin I'm currently having trouble with my google mind reader beta account
20:29 Rakmakallan of course give me a second
20:31 Rakmakallan http://codepad.org/d0C29ZAA      here you go
20:32 deafferret Illegal division by zero at line 77.
20:32 deafferret ?
20:33 Rakmakallan that part is fine because you need to input a filename in the terminal
20:34 deafferret Rakmakallan: is "Illegal division by zero at line 77" an error you are getting and need help with?
20:34 Rakmakallan nope i'm not getting that error at all. Output is fine but results not those expected
20:35 Topic for #bioperlis now http://www.bioperl.org - You can get help here, but you must be patient. Paste your code, input, output, and what you were expecting in a single post on http://gist.github.com/
20:35 deafferret Rakmakallan: I think you need to provide detail about what you were expecting vs. what you're seeing
20:36 deafferret Rakmakallan: and I think we'll need to be able to reproduce your results or it is less likely we will be able to help
20:36 Rakmakallan oh shoot yes
20:36 Rakmakallan i'll send another one with the results
20:39 Rakmakallan http://codepad.org/UMh7kYv1   the output when run in terminal
20:39 Rakmakallan results manually calculated are different. for example first 2 are 0.24 and 0.11 respectively
20:45 deafferret All: For your review: http://bioperl.org/wiki/IRC#Getting_help
20:45 deafferret Rakmakallan: let me look at your new pastes
20:49 deafferret Rakmakallan: is line 4 actually blank?
20:49 Rakmakallan in which paste?
20:50 Rakmakallan in the latter it is
20:50 deafferret AQP1_HUMAN.txt
20:50 deafferret ok
20:51 deafferret Rakmakallan: so here's your code in runnable form, I think  http://github.com/jhannah/san​dbox/tree/master/Rakmakallan/
20:52 deafferret so I assume you're expecting ($tp+$tn+$fp+$fn) to be > 0 ?
20:52 Rakmakallan yes
20:56 deafferret so the proper form of open (EPIC,"$ARGV[0]")    is    open my $epic, '<', $ARGV[0] or die "Can't open '$ARGV[0]'";
20:56 deafferret close $epic;
20:58 Rakmakallan i'm not familiar with this syntax
20:58 deafferret you are now, I just showed it to you.   :)
20:58 deafferret so $SEQ is null
20:59 deafferret does BioPerl have a parser for whatever format AQP1_HUMAN.txt is?
20:59 deafferret (what format is this?)
21:00 pyrimidine TM == transmembrane, maybe?
21:00 deafferret if ($_=~m/^\s{5}(.*)/gi)    matches zero lines in this file
21:01 pyrimidine values are pred. score per residue
21:01 pyrimidine possibly
21:01 Rakmakallan it's transmembrane indeed
21:01 * pyrimidine takes a bow
21:03 pyrimidine and I did that w/o the backscroll, too
21:03 deafferret Rakmakallan: if you put 5 spaces in front of your sequence then your program does stuff
21:04 Rakmakallan what kind of stuff?
21:04 deafferret Rakmakallan: try it and see
21:04 deafferret I have no idea what results you're expecting
21:04 deafferret but the current chain of fail starts with $SEQ being undef
21:05 deafferret Rakmakallan: again, what format is AQP1_HUMAN.txt ?
21:05 deafferret Rakmakallan: you'd be much better off with an actual parser
21:05 Rakmakallan uniprot
21:06 pyrimidine ack, no strictures or warnings?!?  C-style loops?!?  /me eyes burning...
21:06 pyrimidine ah, no, a global filehandle!  Make it stop!
21:06 deafferret pyrimidine: I was gonna go there later  :)
21:06 pyrimidine heh
21:06 deafferret pyrimidine: that correction already in the back-scroll
21:07 Rakmakallan sorry for those, i'm not familiar with programming and it's been 2 months since i started using perl as a biology student
21:07 pyrimidine ah, okay.
21:08 pyrimidine lesson 1: use strict; use warnings;
21:08 pyrimidine lesson 2: don't listen to deafferret, he's insane
21:08 was kicked by deafferret: pyrimidine
21:08 pyrimidine joined #bioperl
21:08 * deafferret laughs evilly
21:08 pyrimidine :-D
21:09 pyrimidine lesson 3: there is no such thing as fight club
21:09 Rakmakallan so rule 1: never talk about fight club doesn't apply
21:10 pyrimidine no, you can talk about it.  We'll just look at you like you're crazy
21:10 Rakmakallan point
21:11 Rakmakallan so anyhow it must either be the double loop or i didn't define the arrays properly
21:17 * pyrimidine looking at it...
21:17 kblin pyrimidine: I write a lot of c code and I don't write loops like that ;)
21:19 pyrimidine :)
21:20 kblin anyway, SIGGF, need to be off. see you tomorrow, folks
21:23 deafferret Rakmakallan: hmm... Bio::SeqIO does lots of file formats, but I don't see 'Uniprot' in the list  http://bioperl.org/wiki/HOWTO:SeqIO
21:23 deafferret did you add 5 spaces and run the program?
21:24 Rakmakallan it's the one listed as "swiss"
21:26 deafferret Rakmakallan: Apparently not. my $in = Bio::SeqIO->new(-file=>$ARGV[0], -format=>'swiss');   doesn't like it
21:27 Rakmakallan apparently you would need the swissprot file to enter its name as an argument
21:27 deafferret Rakmakallan: what? is AQP1_HUMAN.txt a 'swiss' file or not?
21:28 Rakmakallan it is.
21:28 deafferret lol
21:29 deafferret http://github.com/jhannah/san​dbox/tree/master/Rakmakallan/
21:29 * deafferret clocks back in
21:30 Topic for #bioperlis now http://bioperl.org/wiki/IRC#Getting_help
21:38 pyrimidine http://gist.github.com/429231
21:39 deafferret pyrimidine++
21:39 pyrimidine deafferret: that gist has the cleaned up code.  He's getting something undef in the loop, where the 'die' is
21:39 pyrimidine gotta go, car's on the meter and I don't want a ticket
21:39 deafferret shouldn't he be using Bio::SeqIO?   back in ~15m
21:39 pyrimidine yes.
21:39 * deafferret tickets pyrimidine
21:40 pyrimidine ack!
21:40 pyrimidine left #bioperl
21:57 deafferret py[TAB]: requires 5.10   :)
21:58 * deafferret is such a luddite
21:59 deafferret http://github.com/jhannah/san​dbox/tree/master/Rakmakallan/
22:14 nuba joined #bioperl
23:05 CIA-94 biome: Chris Fields master * r4c5a4de / (lib/Biome.pm t/Segment/SegmentContainer.t t/Segment/Split.x): comment out error class stuff, broken in latest Moose 1.07 - http://bit.ly/aF2OiS
23:11 dave_messina joined #bioperl
23:57 russellsmithies joined #bioperl

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