Camelia, the Perl 6 bug

IRC log for #bioperl, 2010-10-01

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

All times shown according to UTC.

Time Nick Message
02:52 melic left #bioperl
02:57 melic joined #bioperl
04:30 buelsbot left #bioperl
06:04 bag_ joined #bioperl
06:24 bag_ left #bioperl
06:40 bag_ joined #bioperl
07:02 bag_ left #bioperl
10:42 carandraug joined #bioperl
10:47 dbolser anyone awake?
10:48 dbolser annoying GFF problem... looks just like my other GFF, but sub-features are not being recognized
10:48 dbolser http://bioperl.pastebin.com/PU4vj1A2
10:49 dbolser http://bioperl.pastebin.com/d0xZqs1d
10:50 dbolser ooooh... strange
10:50 dbolser using +1 and -1 instead of + / - for strand... I wonders...
10:53 dbolser nope... not that... also, search for LuSp239H16 gives no results... /me counts his columns
10:54 dbolser shoot
10:57 dbolser how many informaticists does it take to parse a gff?
10:58 dbolser oh foo
10:59 dbolser looks like you can drop the score column of a gff, and still load it fine, which means my bug is still in force!
10:59 dbolser oops... /me flushes the cache!
10:59 dbolser yay!
11:30 brandi1 joined #bioperl
11:30 brandi1 left #bioperl
12:45 brandi1 joined #bioperl
12:53 brandi1 left #bioperl
13:59 Jun left #bioperl
15:22 deafferret tabs--
15:24 rbuels deafferret: don't speak ill of those tabs.
15:25 rbuels was kicked by deafferret: look out!! that tab is flying straight for your head!! I'll save you!
15:25 rbuels joined #bioperl
15:25 rbuels oh, whew!  it would have gotten me right between the eyes!  thanks!
15:25 deafferret rbuels: you owe me one
15:26 rbuels deafferret++  # saveur aromatique
15:26 deafferret tabs--
15:36 pyrimidine joined #bioperl
15:44 deafferret huh. YYYYMMDDHHMMSS  20101001100110
15:44 deafferret the singularity is approaching
15:50 pyrimidine http://www.genomeweb.com/blog/science-hilarious
15:50 pyrimidine .e.g/ the Ig Nobels have been announced
15:52 deafferret "collect whale snot with remote-controlled helicopters"  ?? cool!
15:53 pyrimidine That's a job I want (controlling the helicopters, not collecting the snot)
16:15 carandraug hi! I've retrieved a sequence with eutils' efetch . I can then use get_response->content or get_response(-file => $path) to print it somewhere. But I was wondering if it's possible to simply load straight into a Seq object.
16:17 rbuels carandraug: yeah, there is a bioperl interface to eutils
16:17 rbuels carandraug: if you use that, things will come out as bioperl objects
16:18 * rbuels looks up what classes they are
16:19 rbuels carandraug: have a look at http://www.bioperl.org/wiki/HOWTO:Getting​_Genomic_Sequences#Using_Bio::DB::EUtilit​ies_to_get_raw_GenBank-formatted_sequence
16:19 rbuels carandraug: and also http://www.bioperl.org/wiki/HOWT​O:EUtilities_Cookbook#esummary_-.3E_efetch , which is linked from that page
16:19 * rbuels wishes his irc client had tinyurl integration
16:20 carandraug rbuels, thanks
16:20 * rbuels recalls too late that a number of bots provide this
16:20 carandraug bots?
16:20 carandraug here on the cahnnel?
16:20 rbuels carandraug: yes there are many bots.   ChanServ, or _ilbot2, or buelsbot
16:20 rbuels carandraug: CIA-95 is also a bot
16:21 carandraug rbuels, can I interact with them and make them questions?
16:21 rbuels carandraug: they do various things ..... ChanServ manages operator privileges for the channel, and some other things .... CIA-95 prints commit messages from bioperl ...
16:21 rbuels i don't think there's a bot on this channel that's interactive ...
16:21 rbuels carandraug: there are some other channels that have them though
16:23 carandraug rbuels, also, I've seen those links. That's from where I got most of the code thus far. But it only prints the sequence. I could save the output and then open with Bio::Seq but I'm trying to do get it straight into an object. Something like Bio::Seq->new($fetcher->get_Response) which doesn't work
16:23 * rbuels looks
16:24 carandraug rbuels, yes. I'm used to the #ubuntu bot, very handy. I wondered if there was something similar here when you mentioned it
16:25 rbuels carandraug: here's some code i use for fetching BAC seqs from genbank http://github.com/solgenomics/tomato_​genome/blob/master/lib/CXGN/TomatoGen​ome/CmdLine/Command/roe_sync.pm#L79
16:26 rbuels carandraug: in that code, see what i do with $gb_full?
16:26 rbuels carandraug: $gb_full->get_Seq_by_acc( 'AC12443' ) will return a Bio::Seq
16:26 rbuels carandraug: (actually a RichSeq, which is a subclass of that)
16:28 carandraug rbuels, hmm.. I'm looking into it
16:30 carandraug rbuels, so I should use the DB::GenBank instead of DB::EUtilities, which would allow me to get everything straight into an object
16:32 rbuels carandraug: yeah i think so
16:32 rbuels carandraug: it would be great if you'd add a section to the wiki about it
16:33 rbuels carandraug: cause obviously if it didn't steer you right, it probably needs some help
16:34 deafferret rbuels++ # wielding his CXGN justice for the good of the people
16:34 deafferret are you completely on github now?
16:34 rbuels deafferret: yep, everything.
16:34 deafferret rbuels += 31
16:35 * rbuels thinks it funny that everybody, including himself, tends to pick prime numbers for things like that
16:35 * deafferret rbuels thinks its cute how scientists notice that numbers are prime
16:35 rbuels at least 7 times out of 11.
16:35 deafferret lol
16:36 * deafferret takes rbuels to vegas as his sekr1t w3ap0n
16:36 rbuels yeah, definitely a prostitute.  yeah....
16:37 deafferret wut?! laugh
16:37 rbuels gotta, gotta count the money.  yeah ...
16:37 deafferret oh   :)
16:37 * rbuels is doing a RAIN MAN IMPRESSION jeez
16:37 deafferret k-mart sucks
16:37 deafferret 10 minutes to wapner
16:37 rbuels ok, enough of this foolishness.
16:37 * rbuels works
16:49 carandraug rbuels, I can't define the region of sequence. By giving only the accession numbers, it get a huge thing, I only want t tiny bit, less than 500bp. Can't I limit it as I did with efetch by setting seq_start and seq_stop? I've looked into http://search.cpan.org/~cjfields​/BioPerl-1.6.1/Bio/DB/GenBank.pm and can't find anything
16:50 rbuels carandraug: yeah, look for the string 'seq_start' in the Bio::DB::GenBank docs
16:51 rbuels carandraug: it's one of those things that's implemented in Bio::DB::GenBank's superclass, but you see it in the synopsis there?
16:51 carandraug rbuels, I see it now. I could sware I didn't saw it a few minutes ago
16:52 * rbuels shakes his head once again at the nightmare of bioperl's inheritance and the way it plays hell with the docs
16:52 deafferret rbuels: don't all inheritance systems have similar doc problems?
16:53 rbuels deafferret: kind of, but some of them actually solve it.
16:53 rbuels deafferret: like javadoc
16:53 * deafferret will need to educate himself
16:53 rbuels deafferret: and doxygen too i think
16:54 rbuels there's the bioperl deobfuscator too, which helps with this a lot
16:54 rbuels carandraug: this is probably the best thing to use for bioperl docs: http://bioperl.org/cgi-bin/deob_interface.cgi
16:55 rbuels carandraug: well, a combination of this and the POD
16:55 carandraug rbuels, deobfuscator. Hoepfully a fitting name
16:56 deafferret ahh... Bio::DB::NCBIHelper    /me nods knowingly
16:57 * pyrimidine is late to the party
16:57 pyrimidine as always
17:01 pyrimidine I likey: http://lmctfy.org/
17:03 deafferret :)
17:06 deafferret so now we need a "let me bioperl deobfuscator that for you" ?
17:09 carandraug rbuels, I finnaly got it working. Thank you very much
17:10 rbuels pyrimidine: yeah, the lm*tfy.orgs are a lovely thing
17:10 rbuels carandraug: no problem, glad to hear it.  pay it forward and help other people.  :-)
17:11 rdesfo joined #bioperl
17:11 carandraug rbuels, I'll add it to the wiki yes. Thanks
17:29 kyanardag joined #bioperl
17:32 rdesfo left #bioperl
17:33 kyanardag left #bioperl
17:47 carandraug rbuels, done http://www.bioperl.org/wiki/HOWTO:Gett​ing_Genomic_Sequences#Using_Bio::DB::E​Utilities_to_get_genomic_coordinates and the small code at the start of http://www.bioperl.org/wiki/HOWTO:Getting_Geno​mic_Sequences#Using_Bio::DB::GenBank_when_you_​have_genomic_coordinates_to_get_a_Seq_object
17:48 carandraug rbuels, the code will look like written by a beginner (it was), but hopefully it's easy to read
17:49 rbuels carandraug++
17:50 pyrimidine carandraug: one of the long-term plans with EUtilities is to have it be the actual user agent for Bio::DB::GenBank, or a new suite of Entrez-related modules.
17:51 pyrimidine Beyond the inheritance from Bio::Root::Root, Bio::DB::EUtilities and ilk could be easily made into a separate distribution
17:52 carandraug pyrimidine, like Ensembl API is a separate distribution?
17:52 pyrimidine sort of
17:53 pyrimidine separately maintained, but available in CPAN
17:54 carandraug pyrimidine, if that happens, I hope somoene from debian with package it. I gave up on ensembl because I couldn't get it working. Thankfully Debian Med is packaging it now
17:55 pyrimidine ensembl is nice, but the API is very particular to the release
17:55 pyrimidine *ensembl release
17:55 pyrimidine not that there's anything wrong with that :)
17:56 carandraug pyrimidine, I was told that with Ensembl I could download the DNA sequence from the gene name or id without going around calculating the coordinates
17:57 pyrimidine yes, you can
17:57 pyrimidine I have done that very thing myself
17:58 pyrimidine carandraug: IIRC the ensembl API page has a few examples of that
17:59 carandraug pyrimidine, I believe you. I looked into example codes and seemed straightforward. I just never managed to install Ensembl. But now got it working using GenBank and EUtilities so it's cool
17:59 pyrimidine ok
18:01 pyrimidine carandraug: if it's any help, I never installed ensembl.  I just have the various perl directories in a specific location, point PERL5LIB at them (along with the latest bioperl), and it plugs along happily
18:05 carandraug pyrimidine, I tried to place them in /usr/share/perl5/Bio/Ensembl/, together with the rest of bioperl but no good. I think the problem may be of network, the university firewall because it kept saying it couldn't connect to their database
18:06 pyrimidine carandraug: have mysql set up?
18:06 pyrimidine if so, you can test the connection directly: http://uswest.ensembl.org/info/data/mysql.html
18:06 carandraug pyrimidine, the mysql module for perl?
18:06 pyrimidine no, just the mysql client program
18:07 pyrimidine but you do need DBI and DBD::mysql for ensembl, IIRC
18:07 pyrimidine *for the ensembl perl API, I mean
18:10 carandraug pyrimidine, yes. I had the mysql client installed and libdbd-mysql-perl
18:10 carandraug pyrimidine, and the link you gave me works fine
18:12 pyrimidine then on the command line run: mysql -h ensembldb.ensembl.org -P 5306 -u anonymous
18:12 pyrimidine carandraug: and then in the mysql shell 'show databases;
18:13 pyrimidine if that connection works, then the API should also work (just need to set the host and port correctly)
18:13 carandraug pyrimidine, I've just found the program. I installed mysql files but not the client
18:13 carandraug s/program/problem
18:14 * pyrimidine is confused
18:14 pyrimidine if the packages are installed, the client program should be available
18:17 carandraug pyrimidine, not exactly. Some program may have needed the mtsql common files but not the clients
18:18 carandraug pyrimidine, let me see why it was installed
18:20 carandraug pyrimidine, the common files are dependency of  libmysqlclient. Many packages depend on that but not on the mysql-client
18:20 pyrimidine true.  on my ubuntu machine, libdbd-mysql-perl depends on libmysqlclient
19:12 carandraug left #bioperl
19:14 arcsine joined #bioperl
20:40 dbolser hello
20:40 dbolser can I be a wiki sysop plz?
20:41 dbolser bioperl is throwing a few 10 warnings in 1000-ish lines, how can I get it to die so I can debug the input?
20:41 dbolser MSG: sorted sublocation array requested but root location doesn't define seq_id (at least one sublocation does!)
20:41 dbolser no idea why a few of my features are throwing this
20:47 deafferret I don't have the key to those keys, sorry
20:47 deafferret but here, have this
20:48 deafferret oops
20:48 deafferret apparently freenode doesn't like "half ops"
20:57 dbolser plz stop
20:57 dbolser oops
20:58 * dbolser goges mad with power
20:58 dbolser was kicked by dbolser: dbolser
20:58 dbolser joined #bioperl
20:58 dbolser well that was embarassing
21:01 dbolser how do I get BP to die when warning so I can debug?
21:07 deafferret a bucket of water?
21:07 deafferret :)
21:09 dbolser ;-)
21:09 dbolser wife home.... night all ;-)
21:09 deafferret I think there's an autodie for that
21:09 dbolser left #bioperl
21:37 pyrimidine left #bioperl
23:00 brandi1 joined #bioperl
23:01 brandi1 left #bioperl
23:36 bag__ joined #bioperl
23:47 brandi1 joined #bioperl
23:47 brandi1 left #bioperl
23:48 bag__ left #bioperl

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