Camelia, the Perl 6 bug

IRC log for #bioperl, 2011-06-13

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

All times shown according to UTC.

Time Nick Message
01:45 carandraug joined #bioperl
02:04 carandraug left #bioperl
02:07 sheenams joined #bioperl
02:31 sheenams left #bioperl
05:44 15SAAQ9SF left #bioperl
08:45 svaksha joined #bioperl
09:29 Tjuh joined #bioperl
09:30 Tjuh hello, is anyone here today?
09:40 dbolser no
09:41 dbolser Tjuh: "while (my $result = $searchio->next_result)" should do it for every hit no?
09:42 Tjuh https://gist.github.com/1021403
09:42 Tjuh tahts what i did
09:42 dbolser Tjuh: yes
09:42 dbolser IRC is a lot like email :-)
09:42 Tjuh and then id like to print a png file for aevery hit
09:42 Tjuh :D
09:42 dbolser "while (my $hit = $result->next_hit)"
09:43 dbolser perhaps you need to move things within that loop?
09:43 dbolser the bit where the PNG gets initalized for example
09:43 Tjuh hmm
09:43 dbolser first get a bp script that does something per hit, then put the PNG generating code there
09:44 Tjuh but if i do that
09:44 Tjuh i wont be able to use $hitname
09:44 dbolser sorry I can't be more specific, busy
09:44 Tjuh to name each png according to its hitname
09:44 dbolser why not?
09:45 dbolser btw, did you search the list of bp scripts for something similar? Seems like it should exist already. If not, perhaps you could submit your result back for inclusion in a bioperl script
09:46 dbolser good luck! P.S. People will come in here as the day goes on and read your questions. It's prolly worth repeating a question once every 12 hours or so if you don't get an answer.
09:53 Tjuh lol ok
09:59 Tjuh left #bioperl
10:09 dbolser so... don't quit
10:35 fy__ joined #bioperl
10:36 fy__ is there a more general bioinformatics irc channel as well?
13:07 Tjuh joined #bioperl
13:07 Tjuh anyone here that has the time to help me?
13:11 dbolser fy__: yes, it's called #bioinformatics
13:12 fy__ hm, that one was created by me dbolser
13:12 dbolser Tjuh: It's better to ask a specific question and wait than ask, "can anyone here help me" (although we all do it)
13:12 Tjuh well i tried
13:13 dbolser fy__: don't judge me !
13:13 Tjuh taht
13:13 Tjuh and no1 asnwered lol
13:13 Tjuh so its better to knwo if some1 is available first
13:13 dbolser Tjuh: AFAIK people like rbuels and pyrimidine (people who are most active in this room) are not out of bet yet
13:13 Tjuh lol!
13:14 dbolser people like kai an kblin should be around, but you have to pique their interest ;-)
13:14 Tjuh its 3:30 pm here
13:14 Tjuh 3:2-
13:14 dbolser sorry to others that I forgot
13:14 Tjuh 20
13:15 dbolser Tjuh: try going into a random chat room and asking 'can anyone help me', and then try 10 times with a precise question that matches the topic of the room... report back on the statistics :-)
13:15 dbolser i.e. do both strategies 10 times
13:16 dbolser fy__: where did you read that I created #bioinformatics?
13:16 dbolser fy__: oh sorry, parsing error
13:17 dbolser fy__: you musta hada typo, because you aren't in #bioinformatics
13:17 fy__ dbbolser: you are right, a typo
13:18 dbolser :-)
13:19 dbolser Tjuh: flu, jhannah and dukeleto may offer suggestions too... ;-)
13:21 Tjuh hmm ok, ill post my question here again: example4/figure4 of the graphics howto shows how to make a png file of only 1 hit, how can i make seperate png files of all the hits in a blast output file?
13:21 Tjuh or is it even possible
13:21 dbolser Tjuh: link to the URL of the howto, and I also point you to my previous advice
13:22 dbolser step 1, make a script that does 'something' per hit, step 2, copy png code there
13:22 Tjuh http://www.bioperl.org/wiki/HOWTO:Graphics starting from Parsing real BLAST ouput
13:23 dbolser cool
13:23 Tjuh i did:
13:23 Tjuh https://gist.github.com/1021403
13:23 Tjuh it does generate a png file
13:23 Tjuh but it is empty
13:24 Tjuh also i closed the previous window, so i dun rmember wut u said
14:18 dbolser Tjuh: nothing specific
15:19 perl_splut Tjuh, it is black because you never add the hit to the graphic
15:20 perl_splut sorry, blank...
15:20 Tjuh ehm
15:21 Tjuh what do u mean?
15:24 perl_splut nvm... missed the $track->add_feature
15:28 Tjuh :S
15:31 perl_splut return unless $feature->has_tag('description'); <-- are you sure each hit gets a tag of description?
15:35 Tjuh one sec
15:36 carandraug joined #bioperl
15:44 Lynx_ Hi all! I don't understand the frame() methods in the Bio::Search hit and hsp objects. I have a fake blast file with just one HSP, but the hit and hsp frame methods give differnt results. How come?
15:45 Lynx_ And the result from hit->frame() is wrong.
15:53 Tjuh perl_splut: lol im so fail, i was using the wrong blastoutput file, i managed to generate a png for each hit in this file: https://gist.github.com/1023037, altho the files are names correctly after the name of the hits, they are all the same
15:53 Tjuh and*
15:54 Tjuh i mean named*
15:55 Tjuh perl_splut: they all look like this: http://imageshack.us/photo/m​y-images/833/2004097170.png
15:55 Tjuh but wit different names
16:02 Tjuh oh wait
16:32 Tjuh ROMA VICTOR :D, i did it lol
16:32 Tjuh thxn for ur help perlslut
16:33 Tjuh oh sry
16:33 Tjuh perl
16:33 Tjuh splut
16:33 Tjuh :D
16:37 Tjuh left #bioperl
17:17 pyrimidine Lynx_: I think you have to stipulate which seq from the HSP from the alignment you want the frame from, hit, query, or list (both hit and query).  Otherwise it always defaults to 'hit'.
17:19 pyrimidine Lynx_: see this bug report regarding the reasons for the change: https://redmine.open-bio.org/issues/2485
17:23 perl_splut I noticed that Bio::Graphics is no longer listed in the Deobfuscator on bioperl.org. Is this due to the split off of Bio::Graphics from Bioperl-live?
17:54 carandraug left #bioperl
18:19 pyrimidine perl_splut: very likely
18:22 perl_splut so, this problem will expand as BioPerl is split up into more subprojects?
18:32 pyrimidine perl_splut: likely, though there are probably ways to work around that (have deobfuscator point at a certain set of blessed Bio* distributions
18:33 pyrimidine or similar)
18:34 perl_splut would hate to see the Deobfuscator die as it has been very helpful with figuring out what methods objects have given the level of inheritance that goes on in BioPerl
18:35 pyrimidine it probably won
18:35 pyrimidine *won't
18:36 pyrimidine all depends on the level of code changes it needs to be updated to deal with the new scheme
18:42 rbuels who maintains the deobfuscator?
18:42 rbuels (anybody?)
18:43 CIA-63 bioperl-live: Chris Fields master * rc73d137 / Bio/SeqUtils.pm : doc patch, fixes bug #3250 - http://bit.ly/mwGQeL
18:44 pyrimidine rbuels: I think it's Dave Messina's
20:09 fy__ left #bioperl
20:39 dnewkirk joined #bioperl
21:12 scottcain joined #bioperl
21:18 scottcain left #bioperl
21:39 scottcain joined #bioperl
22:10 scottcain left #bioperl

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