Camelia, the Perl 6 bug

IRC log for #bioperl, 2010-06-17

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

All times shown according to UTC.

Time Nick Message
01:01 dnewkirk_ joined #bioperl
01:07 brandi joined #bioperl
01:07 brandi left #bioperl
04:08 CIA-52 biome: Chris Fields master * r6218805 / (5 files in 2 dirs): Identifiable using (limited) parameterizable trait - http://bit.ly/be7UZJ
06:06 XaeroOne joined #bioperl
06:21 vinnana joined #bioperl
06:23 vinnana joined #bioperl
07:08 kai durn, as expected, adding the new tests showed that it doesn't work
07:08 kai I hate it when I'm right on these questions
08:05 faceface heh
08:05 faceface so... what has bioinformatics ever done for us?
08:08 vinnana faceface: That's a question that I suspect behind the words of some experimentalists.
08:10 faceface :-P
08:10 faceface did you see the competition?
08:11 kai hah, got it :)
08:12 kai now the HMMer3 parser works for all cases I can think of
08:16 vinnana Which competition?
08:17 faceface 100$ for the best of bioinformatics
08:17 faceface http://kaggle.com/blog/
08:24 vinnana oic
08:25 vinnana hard to answer
08:25 vinnana hard to separate from the contribution the non-bioinformatics contribution
08:26 vinnana maybe blast
08:28 kai genome sequencing?
08:30 vinnana Well, that's an experimental technique.
08:31 vinnana although ... involves a lot a alignments
08:33 faceface I'm guessing you can argue tools
08:33 faceface i.e. 'genome alignment' as a tool
08:33 faceface citing the relevant biology that it enables
08:34 faceface I'm definately thinking about writing blast as an entry
08:34 faceface There can be more than one entry on each topic
08:35 kai vinnana: arguably. but I can manually do alignments for small genes or proteins
08:36 kai there's no way you could sequence a genome and manually do all the read merging and be done before your project funding runs out :)
08:37 vinnana there is no way to search a sequenced cDNA in refseq without blast (or an equivalent tool, but blast is one of the first)
08:37 kai vinnana: right. blast probably is one of the most used bioinformatics tools
08:37 vinnana kai: At Sanger's times sequencing was pretty manual.
08:39 kai vinnana: sure, but in new projects, you don't get funding if you don't do "metagenomics"
08:39 kai if it's not at least 10 organisms, it's boring
08:40 vinnana Hey I see: My answer is besides the question. The question is "what is the single greatest biological discovery made possible by bioinformatics?"
08:43 vinnana kai: You mean, when you do sequencing?
08:44 kai when you try to get grant funding for projects that include sequencing
08:48 vinnana What about genotyping populations (one by one) or sampling [sub]species?
08:51 kai well, I guess you don't have to sequence them in one go, but you need to sequence more than one organism :)
08:53 vinnana I have to admit, there is no "single greates biological discovery made possible by bioinformatics" that's obvious to me.
08:54 kai it's a hard question
08:55 kai especially as you could argue that most discoveries would be possible even without bioinformatics, they would just take longer
08:55 kai see our discussion about genome sequencing
08:56 vinnana ... possible in a reasonable amount of time?
08:57 vinnana Just, leaving away the 'bioinformatics' -- how to judge whether a biological discovery is the greatest?
08:57 kai well, in cases like drug design, how do you factor that? you could just burn a couple of more millions and do a broader set of screening experiments
08:58 kai that's a good point as well :)
09:00 kai hehe
09:01 kai I should have brought some pop-corn
09:01 vinnana :-)
09:03 kai I just asked my co-workers on a coffee break, so now they're arguing about the greatness of biological discoveries
09:03 vinnana kai: Are you working in Tuebingen?
09:08 kai yeah
12:50 brandi joined #bioperl
12:55 brandi left #bioperl
13:24 faceface I still think blast is a valid answer
13:25 faceface because of the range of biology that 'would not have been *practically* possible' without such a tool.
13:25 faceface of course, nothing done by bioinformatics is logically impossible without it
13:25 faceface perhaps none of it is even physically impossible
13:26 faceface but it's practically impossible to find and align all the globin sequences in genbank without tools like blast
13:26 * faceface goes for 100$
13:26 faceface wait... I always find it confusing that you say it "one hundered dollars" but it "$100"
13:27 faceface but write it "$100"
13:31 vinnana faceface: but blast is not a biological discovery ...
13:32 faceface no, but the discovery of mutation paterns in teh structure of heamoglobin is...
13:33 faceface actually, max-perutz did that by hand
13:33 faceface JMB somewhere
13:37 faceface how about you argue that collectively, all the biological discoveries made by blast add up to a big contribution to the field?
13:37 vinnana maybe like this
13:37 vinnana findBioinfoDiscovery(discovery) {
13:37 vinnana if (no_chance_without_bioinformatics(discovery)) {
13:37 vinnana return discovery;
13:37 vinnana } else {
13:37 vinnana findBioinfoDiscovery(nextGr​eatestBiologicalDiscovery);
13:37 vinnana }
13:37 vinnana }
13:37 vinnana findBioinfoDiscovery(theReally​GreatestBiologicalDiscovery);
13:38 vinnana ;-)
13:39 faceface it would be neater in prolog ;-)
13:40 vinnana sure ... but that's a bit too long ago for me (~8-9 years)
13:46 kai scheme? :)
13:50 CIA-52 biome: Chris Fields master * r4d65ecf / (lib/Biome/Role/Annotate.pm lib/Biome/Role/Identifiable.pm): meta role not working yet for parameterized roles - http://bit.ly/bguZb0
13:56 vinnana common lisp? (labels ((find-discovery (discovery) (if (no-chance-without-bioinfo discovery) discovery (find-discovery (next-greatest-discovery))) (find-discovery +greatest-biological-discovery-ever+))
13:57 vinnana Maybe: The great importance of small RNAs is quite significant and many small RNAs are found using bioinformatics.
14:34 faceface vinnana: I'm not sure how the comp will be judged, but you can enter multiple times... you should enter the code solutions
14:35 * faceface doesn't like sRNA personally ...
14:35 vinnana :-)
14:41 vinnana left #bioperl
19:04 deafferret ..o.o__..
19:06 deafferret has bioinformatics contributed to human health in any way?
19:11 perl_splut good question, but not one I have an answer for, heheh
19:57 bag_ joined #bioperl
20:06 kblin evening folks
20:07 * deafferret waves
20:36 Jun joined #bioperl
21:27 CIA-52 bioperl-live: Kai Blin master * r7402205 / (3 files in 3 dirs):
21:27 CIA-52 bioperl-live: Bio::SearchIO::hmmer3: Fix parsing hmmscan results with multiple domains
21:27 CIA-52 bioperl-live: The fix to support hmmer results without alignments broke the parser for
21:27 CIA-52 bioperl-live: results with multiple alignments per hit. Rework the code to support
21:27 CIA-52 bioperl-live: this. (+11 more commits...) - http://bit.ly/9g2vTD
21:28 kblin whee
21:32 perl_splut heheh
21:41 dnewkirk_ joined #bioperl
21:48 deafferret kblin++
21:48 deafferret all done?
21:49 deafferret kblin: now add all that stuff to biome too, would ya?  ;)
21:53 * deafferret wonders if py[TAB] has considered biome using bioperl-live for stuff-not-moosey-yet or stuff-where-moose-doesnt-help
21:56 rbuels hmm.  he should probably just change his handle to pytab.
21:56 * deafferret summons py[TAB]
21:56 deafferret pyreallyhardtospellimine
21:56 rbuels whiner.
21:56 deafferret traitor
21:57 deafferret i'm used to all these other blokes ditching me at YAPC. your ditching me is hurtful
21:58 rbuels :-(
21:59 rbuels i'm ditching because i have had *so* much crap going on lately, and i am desperately looking for things that i can cut without being a.) fired or b.) disowned by my family
22:05 deafferret oh. sorry. I'll shut up then
22:05 deafferret (on phone)
22:05 * deafferret hugs rbuels
22:06 deafferret i think I'll take a cardboard rbuels cutout with me to YAPC
22:06 deafferret "what do you think, rob?"
22:06 deafferret "really? you like that idea? cool"
22:06 deafferret "corn is a fucker you say? i see."
22:11 * deafferret picks the cardboard cutoff back off the floor, tapes it into the chair
22:11 deafferret cutout
22:11 deafferret STAY
22:21 deafferret rbuels: wish I could help, sorry...  :(
22:21 deafferret ooo (bowtie-bio.sourceforge.net)++
22:22 deafferret 17 hours to build an index or 20m to download one they already built for me? hmm... tough choice here
22:39 deafferret ha... getting support request for random software from people in India...  http://biodoc.ist.unomaha.edu/​rt/Ticket/Display.html?id=1085  guest guest
22:41 rbuels deafferret: actually, i kind of need that manual too.  could you send it over?  kthx
22:42 * deafferret autodocks rbuels
22:42 deafferret oh, oops, wrong ticket
22:42 * rbuels coldcocks deafferret
22:42 * deafferret accepts lovingly
22:42 rbuels gah
22:43 deafferret oo, that is cold
23:04 deafferret toodles. off to $work[2] for some freelancing   :/
23:04 deafferret &
23:52 spekki0 joined #bioperl
23:53 spekki0 hey guys i keep getting this error http://codepad.org/9R7KTVP3 is this some issue with ncbi's servers or?

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