Camelia, the Perl 6 bug

IRC log for #bioperl, 2009-08-27

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

All times shown according to UTC.

Time Nick Message
03:41 kyanardag joined #bioperl
12:21 kyanardag joined #bioperl
12:36 faceface rbuels: ty
13:34 kyanardag joined #bioperl
13:52 kyanardag_ joined #bioperl
14:16 jorvis_ joined #bioperl
14:17 jorvis_ Do you know of a way, with Bioperl, to extract a subsequence from an annotated Seq feature and have the positions of the annotations on it recoordinated based on the subsequence?
15:08 deafferret jorvis_: I'm not sure what recoordinated means. I'm no annotation expert.  :)
15:09 jorvis_ I just mean that if a gene has coordinates 100-200 on some contig, and I somehow pull a subsequence of that contig with its associated annotation from bases 50-300, I'd get a new feature with the same gene at coordinates 50-150
15:11 deafferret ah, yup. You're past my annotation depth, sorry
16:56 faceface left #bioperl
17:27 xp_prg jorvis_ explain to use more about annotations!
17:30 rbuels jorvis_: i'm not aware of existing code in bioperl to transfer annotations like that.....
17:31 rbuels jorvis_: i'm looking around for it.....
17:31 rbuels jorvis_: you should probably send a question on it to the mailing list
17:35 rbuels aha, it sounds like jason stajich started working on this very thing....
17:35 rbuels but it's not finished yet
17:38 rbuels ok, i sent a message about it to bioperl-l
17:40 rbuels looks like bug 1572 is related to this
17:41 rbuels http://bugzilla.open-bio.org/show_bug.cgi?id=1572
17:42 xp_prg what is an annotation?
17:43 rbuels xp_prg: an annotation is a region on a sequence that is interesting for some reason
17:44 xp_prg hmm... ok
17:44 rbuels xp_prg: if a sequence has matches to some other sequence, you can put an annotation on it saying 'this matches to that other thing'
17:44 rbuels xp_prg: or if you run a gene finder on a sequence, you can put annotations on it for where the gene finder thought it found something
17:45 xp_prg so it is like a bookmark
17:46 rbuels well....not really
17:46 rbuels it's just another type of data
17:46 xp_prg does it not hold information about position in the sequence?
17:46 rbuels well, yes i guess it does.
17:48 xp_prg rbuels I am having my weekly diybio meeting at www.justin.tv/xp_prg tonight at 8 pm pdt, I hope you will join us :>
17:49 rbuels what is that?
17:49 xp_prg it is a do it yourself biology group, we are like a linux group etc...
17:49 xp_prg we present, teach, and discuss synthetic biology etc...
17:50 xp_prg we are going to do a gfp experiment Saturday after this one
17:50 xp_prg rbuels where do you live just curious?
17:51 rbuels portland, or
17:51 rbuels i work for sgn.cornell.edu
17:51 xp_prg oh awsome
17:51 xp_prg perhaps you would be interested in speaking at a future meeting via justin.tv remotely
17:51 xp_prg we would really appreciated it!
17:53 rbuels heh, well thursday nights are pdx hackathon: http://calagator.org/events/1250457617
17:54 rbuels diybio is an SF group?
17:54 xp_prg it has an SF group if you will
17:54 xp_prg rbuels why not present from there one night?
17:54 xp_prg it could be as little as 10 to 20 minutes, we don't care!
17:54 xp_prg someone with your knowledge and experience would be greatly appreciated
17:55 xp_prg what do you work on at this pdx hackathon?
17:55 rbuels heh, mostly drinking lately
17:55 rbuels sometimes the bioperl gff refactoring
17:55 xp_prg well take 10 to 20 minutes and present to us
17:55 rbuels sometimes eulerbench
17:55 xp_prg you could even do it tonight if you want :>
17:55 xp_prg do you have a laptop?
17:56 rbuels heh yes of course, but does justin.tv's client stuff work on linux?
17:57 xp_prg I haven't tried it, I think so, of course you could choose a different technology if you like
17:57 xp_prg that has been easy for me to use
18:00 * rbuels looks at the diybio meeting recordins
18:05 rbuels that's funny, james clements is from ithaca
18:05 rbuels or was there for a while
18:06 rbuels if he owned the chapter house and the haunt
18:06 rbuels i used to go to those all the time, lol
18:07 xp_prg oh cool!
18:07 xp_prg what do you say, want to present tonight?
18:07 rbuels heh, not tonight, but maybe later
18:08 xp_prg thanks, we are looking for speakers so ask your other bio friends too if you don't mind :>
18:08 rbuels i'm not sure if the stuff i know about would be particularly interesting to your clientele.
18:08 xp_prg it is trust me
18:08 rbuels heh ok
18:09 rbuels what i do, is i run the annotation and data handling pipelines for the tomato genome project
18:09 xp_prg dude that is so amazingly interesting to us
18:09 rbuels so i'm more of a software engineer than a biologist.
18:09 xp_prg so am I
18:09 xp_prg I am not a biologist, the fact that your more of a software engineer is an extra boost!
18:09 rbuels so are you Tim?
18:10 * rbuels is watching the intros in the meeting recording
18:11 xp_prg yup :>
18:12 rbuels so what is your software experience?
18:12 rbuels are you a perl coder?
18:12 xp_prg much experience, CSE, MSCS, MBA
18:13 xp_prg score 99% on a perl test for a potential job
18:13 xp_prg went to the perl meetings in Santa Rosa all the time
18:14 xp_prg ask me perl questions if you don't believe me :>
18:14 xp_prg nothing like a blessed reference to get your oo on ;>
18:15 rbuels lol i believe you
18:15 rbuels well bioperl is always looking for help
18:16 rbuels if you are looking around and the docs don't make sense, it's probably cause they suck and need help ;-)
18:17 xp_prg awsome
18:19 xp_prg I know perl was used to knit together the first human genome sequencing, do you know how that code worked?
18:19 rbuels no, i'm not familiar with it
18:19 rbuels maybe it's archived somewhere
18:20 rbuels it's likely to be pretty horrible stuff, since it was such a fast job
18:20 rbuels horrible stuff == get the data out the door right now!
18:20 xp_prg I am not sure it was opensourced
18:20 xp_prg it would be cool to understand it
18:20 rbuels well, wgs_assembler is open source i think
18:20 xp_prg so do you like the videos of our diybio meetings?
18:20 rbuels which is the child of the assembler Celera wrote
18:21 rbuels yeah i think it's a good idea to do the video thing
18:33 xp_prg rbuels tell me more of what you do at your job, how do you work with the data, how does it come in, etc... are you using microarrays for sequencing?
18:34 rbuels microarrays aren't really used for genome sequencing, they're a tool for doing expression analysis
18:34 rbuels but they're somewhat obsolete now with the advent of next-generation sequencing technologies
18:34 xp_prg oh wow tell me more please!
18:34 rbuels because you can use next-gen sequencers to accomplish the same thing
18:35 rbuels heh, this isn't my area really, i'm just regurgitating (probably incorrectly) the things i've heard from others
18:36 rbuels but microarrays involve spotting oligonucleotides on a chip
18:36 rbuels and then squirting your liquid sample on the chip surface
18:36 rbuels and the oligos are specially formulated to glow in various ways when they hybridize to dna/rna in the sample
18:36 rbuels usually rna i think
18:36 rbuels and then you read off the glow pattern with a ccd camers
18:37 rbuels s/s$/a/
18:37 rbuels so you can tell the relative abundance of different rna in the sample
18:37 rbuels and the sample would presumably have been taken from some kind of tissue you're interested in
18:37 rbuels like a mouse liver, or a plant bud at exactly 2 weeks of age, or something
18:38 rbuels which should presumably give you some clues as to which genes are being expressed and their amounts in the tissue you sampled
18:38 rbuels so i guess that's microarray analysis in a nutshell
18:38 rbuels a very small nutshell.
18:38 rbuels lol
18:39 rbuels problem is, the microarray data is really noisy
18:39 rbuels and tricky to work with
18:39 rbuels lots of hairy statistics that people often don't do correctly
18:39 rbuels so a lot of people are really suspicious of all the microarray data they see
18:39 rbuels with next-gen sequencing though,
18:40 rbuels you can just sequence your sample, and count the relative abundance of the different patterns in the reads you get
18:40 rbuels and there you have it, you've just measured the relative abundance of different rnas in your sample
18:40 rbuels more or less
18:40 rbuels the details of all of this are deep and complicated and I don't know about them.
18:41 rbuels lol
18:46 rbuels xp_prg: you ought to subscribe to bioperl-l also
18:46 rbuels if you haven't already
18:48 xp_prg yes I should, tell me more of what you do with bioperl in your job if you don't mind
18:50 rbuels well, nominally i handle tomato genome sequence data that comes in from the international tomato genome sequencing project (see http://sgn.cornell.edu/about/tomato_sequencing.pl)
18:51 rbuels and i designed and built the annotation pipelines for the tomato genome pieces that are coming in
18:51 rbuels i use bioperl mostly for parsing fasta-format sequence files
18:51 rbuels and for parsing and writing various annotation formats
18:51 rbuels mostly GFF3 format for annotations
18:51 xp_prg hmm... interesting
18:51 rbuels see http://song.sourceforge.net/gff3.shtml
18:51 rbuels for the gff3 spec
18:52 rbuels right now, i'm working on an overhaul of bioperl's annotation (a.k.a. feature) handling systems
18:52 xp_prg oh way cool
18:53 rbuels to make a strongly-typed annotation object that is tied to one or more ontologies
18:53 rbuels ontologies are a big thing in bio
18:53 xp_prg what is an ontologie?
18:54 philsf joined #bioperl
18:54 rbuels basically one way of encoding knowledge in machine-readable form
18:55 * rbuels looks for a reference to point xp_prg at
18:55 xp_prg you see why presenting would be good for our group?
18:55 rbuels http://www.geneontology.org/
18:55 rbuels yeah, another name for an ontology is a "controlled vocabulary"
18:56 rbuels in which the terms have relations to one another
18:56 rbuels that are (always, i think) encoded as a tree
18:56 rbuels a.k.a. directed acyclic graph
18:57 rbuels mostly the ontology terms that i personally worry about are the ones in the SO
18:57 rbuels http://song.sourceforge.net/so.shtml
18:57 rbuels this is a controlled vocabulary for describing features/annotations on sequences
18:58 rbuels the GFF3 format specifies that features need to be described using ontologies
18:58 rbuels in contrast to before, when the descriptions were just free text
18:58 xp_prg rbuels what is your opinion of biopython?
18:58 rbuels i don't really know enough about python to have one.
18:59 rbuels in my forays into python, it's struck me as clean, but ultimately limiting.  but that might just be because i don't know it deeply.
19:00 rbuels just like people who say perl is unmaintainable just say that out of ignorance.
19:00 * rbuels shrugs
19:00 rbuels perl is pretty deeply entrenched in bioinformatics i think
19:01 rbuels one thing i think the wider world (outside of academia) could really benefit from is for people to know when they don't know what they're talking about
19:02 rbuels that's one thing academics are good at.  saying "i don't know what i'm talking about"
19:02 rbuels lol
19:02 rbuels anyway, this is just rambling and i need to get back to work
19:03 xp_prg ok, thanks again for talking with me, I hope you will join us tonight at www.justin.tv/xp_prg at 8 pm pdt
19:03 xp_prg it is passive you don't have to participate if you don't want to
19:03 rbuels probably not, i just got back in town, and i promised to cook a nice dinner for the wife tonight
19:03 rbuels :-)
19:04 rbuels but good luck with it
19:04 rbuels keep hanging out in here, and encourage others to
19:04 rbuels also #gmod on the same server is the channel from gmod-related stuff
19:04 rbuels see http://gmod.org
19:05 rbuels #gmod as a channel is just getting started
19:05 rbuels it doesn't have a registration with freenode yet, but i've put in for one
19:06 xp_prg what is gmod about?
19:07 rbuels making open-source components for building MODs
19:07 rbuels sgn.cornell.edu is a mod
19:07 rbuels as is flybase.org
19:07 rbuels wormbase.org
19:07 rbuels yeastgenome.org
19:07 rbuels chlamybase.org
19:07 rbuels dictybase.org
19:08 rbuels gramene.org
19:08 rbuels MOD = model organism database
19:08 rbuels sort of the home on the web for all the stuff that's known about the genetics and genomics and sometimes phylogenetics related to organism X
19:09 xp_prg wow!!!!
19:09 rbuels human doesn't have a MOD, probably because there are so many people involved in it and they can't agree
19:09 rbuels but i don't really know, not being a part of the human genome community
19:09 rbuels for human, there's hapmap.org....
19:09 xp_prg what is in a model organism database?
19:10 rbuels erm, usually the genome sequence, and annotations
19:10 rbuels and some kind of database of genes in that organism
19:10 rbuels ESTs usually
19:10 xp_prg what is an EST?
19:10 rbuels EST=expressed sequence tag
19:10 * rbuels finds ref
19:10 rbuels essentially the sequence of expressed RNA
19:11 rbuels http://en.wikipedia.org/wi​ki/Expressed_sequence_tag
19:11 xp_prg how might someone use these mods?
19:11 xp_prg by the way this is exciting!
19:11 rbuels RNA reverse-transcribed to cDNA, and then sequence that cDNA
19:11 rbuels well, people researching pathways in c. elegans look up stuff on wormbase.org
19:11 jorvis_ rbuels: thanks for the info (I was afk a bit)
19:11 rbuels like, i dunno....
19:11 rbuels oh there he is
19:11 xp_prg research pathways to accomplish what kind of thing?
19:12 rbuels yes, there is a nice thread on bioperl-l about your stuff now jorvis_
19:12 xp_prg I know I keep asking questions, but still learning :>
19:12 rbuels have a look
19:12 jorvis_ oh - started by you?  :)
19:12 rbuels xp_prg: yes of course
19:12 rbuels jorvis_: yes, lol
19:12 rbuels jorvis_: got responses from both chris and jason now though
19:12 jorvis_ looking now
19:12 rbuels jorvis_: not bad for a couple horus
19:12 jorvis_ I'm on that list - just have it auto-filtered in Gmail
19:12 rbuels xp_prg: let's seeeee
19:13 rbuels xp_prg: well i can give an example from my wife's work
19:13 rbuels xp_prg: she's a specialist at squirting colorless liquids from one place to another
19:13 rbuels xp_prg: a.k.a. a regular biologist
19:14 jorvis_ I can weigh in there then - Jason knows me
19:14 rbuels xp_prg: she works on the chemical interactions involved in asthma reactions
19:14 rbuels jorvis_: are you a fungus man?
19:14 jorvis_ that's a terrible thing to say
19:14 jorvis_ (hehe)
19:14 rbuels lol
19:14 jorvis_ but no, general annotation
19:14 rbuels jorvis_: ah.
19:14 jorvis_ I worked at TIGR/JCVI and now at IGS
19:15 rbuels xp_prg: my wife is studying interactions involving chemokines and receptors and all that jazz
19:15 jorvis_ Just know Jason through many meetings, etc.
19:15 rbuels xp_prg: and a receptor is just another kind of protein (as i understand it)
19:15 rbuels xp_prg: so when she wants to run some kind of experiment of something in mouse, for example
19:15 rbuels xp_prg: and she needs the protein or genome sequence for some gene
19:15 rbuels xp_prg: in order to set up a PCR or something
19:16 rbuels xp_prg: she goes to that mouse mod....
19:16 rbuels xp_prg: what's it called
19:16 * rbuels looks
19:16 rbuels mgi
19:16 rbuels http://www.informatics.jax.org/
19:21 xp_prg thanks so much for this help
19:21 rbuels no prob.  i could give a short talk on one or more of these topics sometime.
19:21 rbuels it would help if you'd kind of pick the topic from among all this stuff  :-)
19:22 xp_prg well I am mostly interested in synthetic biology
19:22 xp_prg specifically bio bricks and creating bio bricks
19:24 rbuels well i don't really know anything about that  :-)
19:31 vein joined #bioperl
19:43 xp_prg ok :>
20:36 deafferret "squirting colorless liquids from one place to another"  lol
20:48 rbuels deafferret: i aim to please
20:49 rbuels deafferret: i'll be here all week.  tip your waitress.
22:28 rbuels seen pyrimidine?
22:28 rbuels blech.  no useful bots.
22:42 vein joined #bioperl
22:47 kyanardag_ joined #bioperl
22:59 ptl joined #bioperl
23:52 xp_prg there is a meeting tonight at the tech shop for the diybio group starting at 8 pm
23:52 xp_prg keep it real or watch at www.justin.tv/xp_prg
23:55 jorvis_ joined #bioperl

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