Camelia, the Perl 6 bug

IRC log for #bioperl, 2011-01-21

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

All times shown according to UTC.

Time Nick Message
00:04 brandi joined #bioperl
00:05 brandi left #bioperl
00:09 philsf left #bioperl
00:15 brandi joined #bioperl
00:15 brandi left #bioperl
02:41 melic left #bioperl
05:54 dukeleto left #bioperl
05:55 dukeleto joined #bioperl
06:52 bag joined #bioperl
07:59 bag left #bioperl
08:31 Vutral left #bioperl
08:43 Vutral joined #bioperl
12:50 Vutral_ joined #bioperl
13:56 brandi joined #bioperl
13:58 brandi left #bioperl
16:19 Dezin joined #bioperl
16:19 Dezin anyone here?
16:32 ank joined #bioperl
16:37 Dezin anyone?
16:58 splut nope, no one
16:58 splut left #bioperl
16:58 perl_splut joined #bioperl
16:59 Dezin lol
16:59 Dezin so ill just post this
16:59 Dezin to no one
17:00 Dezin i had a problem with efetch
17:00 Dezin my $factory = Bio::DB::EUtilities->new(-eutil => 'efetch',                 -db => 'protein', -rettype => 'genbank', -id => $_, -email => 'joshpk105@yahoo.com');         $factory->get_Response(-file => $_);
17:00 Dezin i did that
17:00 Dezin but it didn't save the file in genbank format
17:00 Dezin it was something else
17:06 rbuels Dezin: if that is in a loop, try using a named variable instead of $_, does that help?
17:06 perl_splut $_ is fine in a loop depending on what in the loop is setting it, heheh
17:06 rbuels Dezin: and have you read the howto documents on the bioperl wiki?
17:06 Dezin yeah
17:06 rbuels perl_splut: not if the code that you're calling inside the loop is clobbering $_
17:07 rbuels perl_splut: which happens all too often
17:07 Dezin ill test with hard code
17:07 rbuels perl_splut: Bio::Graphics for example had lots of routines that clobbered $_ until i fixed them
17:07 perl_splut guess I never ran into that
17:08 rbuels Dezin: if you keep having problems, use http://gist.github.com and put your full script there
17:09 rbuels Dezin: so other people can run it and see what is going on
17:09 Dezin yeah
17:09 Dezin is printed say result
17:09 Dezin *same
17:09 Dezin k
17:10 perl_splut what is it returning if not genbank?
17:10 Dezin i have no idea
17:10 Dezin ill paste some
17:10 Dezin Seq-entry ::= set {   class nuc-prot ,   descr {     source {       genome genomic ,       origin natural ,       org {         taxname "Amsacta moorei entomopoxvirus 'L'" ,         db {           {             db "taxon" ,             tag               id 28321 } } ,         orgname {           name             virus "Amsacta moorei entomopoxvirus 'L'" ,           mod {             {               subtype old-name ,               su
17:10 perl_splut ahh, it is printing the hash
17:11 perl_splut or something like that
17:12 Dezin should i really create this gist thing?
17:12 perl_splut nvm. That is ASN format
17:13 Dezin git://gist.github.com/790012.git
17:14 perl_splut so, sounds like either genbank format isn't available for that protein or the flag isn't being properly set to get it
17:14 Dezin its a genome and it is available
17:15 Dezin i just d/led it
17:15 perl_splut you aren't grabbing a genome. You requested it search the protein database
17:15 Dezin yeah i started off on the genome db
17:15 Dezin didnt work
17:15 Dezin then i tried nucleotide
17:15 Dezin and the same thing happened
17:15 Dezin as protein
17:18 perl_splut did you try rettype => 'gb' ?
17:18 perl_splut instead of rettype=>'genbank'
17:19 Dezin i didnt
17:19 Dezin i will
17:19 perl_splut http://www.bioperl.org/wiki/HOWTO:EUtilities_C​ookbook#Retrieve_raw_data_records_from_GenBank.2C_save_raw_data_to_file.2​C_then_parse_via_Bio::SeqIO
17:19 perl_splut shows that rettype is gb not genbank
17:20 Dezin hmm it did work
17:20 perl_splut so, the default must be ASN if either BioPerl or Eutils doesn't recognize the type requested
17:20 Dezin ah
17:20 Dezin thx
17:21 Dezin heres a question
17:21 Dezin if i have more than one genbank record in the same file
17:21 Dezin can i iterate through them with seqio?
17:21 perl_splut can try
17:22 Dezin lol
17:22 perl_splut while (my $seq = $seqin->next_seq) {
17:22 perl_splut # do whatever....
17:22 perl_splut }
17:22 Dezin yeah i know
17:23 Dezin doesnt seem like it
17:23 rbuels i would be surprised if you couldn't
17:23 rbuels Dezin: just make sure you're using the right seqio
17:23 Dezin true
17:23 Dezin is it genank or gb
17:23 Dezin *genbank
17:24 rbuels Dezin: 'genbank'
17:24 perl_splut just like it shows in the example in the docs I linked to :)
17:25 Dezin yeah
17:25 Dezin i swear i have read that document multiple times
17:25 perl_splut The devil is in the details
17:25 perl_splut as are the bugs
17:25 Dezin truth
17:26 Dezin soooo it looks like this doesnt work
17:26 Dezin should i show you guys or just do it another way?
17:26 perl_splut save each record to its own file then
17:26 Dezin yeah
17:26 Dezin thats what i meant
17:27 Dezin thx for the help
17:36 Dezin left #bioperl
17:39 pyrimidine joined #bioperl
17:42 pyrimidine ah, pity, Dezin left
17:43 perl_splut his issue was resolved
17:44 pyrimidine seemed like he had trouble parsing the eutils output
17:44 perl_splut he was using the wrong rettype
17:44 pyrimidine hard to say w/o seeing what he got
17:44 perl_splut he wanted genbank, but was getting ASN
17:45 perl_splut because he had set the rettype to 'genbank' instead of 'gb'
17:45 perl_splut not sure if it is BioPerl or Eutils that ignores a rettype that isn't recognized
17:45 pyrimidine he mentioned this after that, though: http://irclog.perlgeek.de/b​ioperl/2011-01-21#i_3207872
17:46 pyrimidine perl_splut: it's a bit of both
17:46 perl_splut not sure what message you were pointing at. Guessing it was the multiple Genbank records in a single file
17:47 pyrimidine 17:26:<Dezin>: soooo it looks like this doesnt work
17:48 pyrimidine but no big deal
17:48 perl_splut ok, that is in reference to multiple Genbank records in a single file
17:48 perl_splut looping with SeqIO didn't work for him I guess
17:49 pyrimidine yeah, that's what I was curious about, makes me wonder if the format has changed and foobar'ed the parser
17:49 Vutral left #bioperl
17:49 pyrimidine easy enough to check
17:53 pyrimidine seems to be working fine in the test suite
17:54 Vutral_ left #bioperl
17:54 perl_splut don't know what version of BioPerl he was using or if his own code was clobbering the records
18:09 rbuels ew.  ignoring an invalid rettype is a very bad behavior.
18:09 Vutral joined #bioperl
18:26 Vutral_ joined #bioperl
18:30 pyrimidine rbuels: problem is, NCBI has made these a bit of a moving target, changing some of the values, not documenting them all, aliasing them, etc.
18:30 pyrimidine rbuels: I had some error checking code in during one iteration of development and removed it after they made too many changes.  PITA
18:30 * rbuels sighs
18:31 * rbuels hums his "software is for suckers" theme song
18:31 pyrimidine :)
20:02 deafferret candy? where?
20:15 deafferret wow. spam got through my 2 layers. impressive
20:15 deafferret pretty rare that gmail + Mail.app doesn't do the trick
20:15 deafferret and/or  pobox.com + Mail.app
20:15 dukeleto left #bioperl
20:16 dukeleto joined #bioperl
21:55 Dezin joined #bioperl
23:06 Vutral_ left #bioperl
23:07 Vutral_ joined #bioperl

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