I thought about doing a colony counter for the G1 ... point it a plate , click and get a the number of colonies. This could be handy :)
- Pedro Beltrao
Mobile versions of commonly-used websites: NCBI databases, BLAST, PDB etc?
- Neil Saunders
Like Pedro's idea too. A common complaint about electronic lab notebooks is that people in the lab do not record data at a computer - so why not transmit to the ELN via phone? Better still, have the instruments tweet to the ELN :-)
- Neil Saunders
Would there have to be one? (Or am I just narrow-minded here? Hope not...) The colony counter makes sense, but there's no real added value in blast or ncbi on a mobile phone, is there? It'd really have to be an app that leverages the fact that it can be carried around. So maybe as an input portal for your labjournal (notes, photos).
- Jan Aerts
I don't know about bioinformatics, but biology could sure benefit from a killer lab timer. :-)
- Danielle Fong
Would any aspect of bioinformatics benefit from being made mobile at all? Maybe a service that could check in on the progress of molecular dynamics calculations or something?
- Mr. Gunn
Yes, it has to be said that most bioinformaticians are rarely away from a computer long enough to require mobile access :-)
- Neil Saunders
@Pawel nice link. I will check it out.
- Paulo Nuin
@Neil: may I guess where you have lunch?
- Paulo Nuin
the monitoring visualization bits are relatively east ESP with a full featured browser. I love the colony idea. In a lab finding a sample by taking a picture of the bar code (for the g1) or using the phone as some sort of controller
- Deepak Singh
@Paulo What? You don't eat your lunch at the computer?
- Marcos de Carvalho
On Wednesday mornings (GMT+2) we have a bioinformatics lab with ca. 40 students. From our experience, NCBI team chose exactly these moments for introduction of the improvements, because every second week the whole site is broken for half an hour :).
- Pawel Szczesny
Error level (near 0%) for SwissProt looks interesting. People of protein-protein interaction data claim 2-9% error rate on manually curated sets, and the same level I would expect from SP. Some things are better than I thought ;).
- Pawel Szczesny
Not too bad, but I think its even worse than they do. And evidence codes won't fix this. Manual curation and standards for automatic annotation are in dire need of a revolution, and even if we get that it'll still take years to fix.
- Paul J. Davis
@Paul How would you revolutionise it if all the data is still contained in the relatively unnaccessible journal article?
- Frank
Trust Pawel to see the half full (or 60% full) part of the glass. I agree with Frank though: it is not feasible to go through NR and fix the annotations manually. We just have to accept that NR, TrEMBL and KEGG are (mostly) over-annotating, and remember that when we rely on them when delving into the protein family level.
- Iddo Friedberg
Thanks Iddo. I lost count of how many times annotation errors came up in my discussion with experimentalists who lack experience with such databases. (Not surprisingly, they usually think these errors are negligible, especially when it comes to THEIR proteins.) Now I'll just send them a link to your post...
- Mickey Kosloff
yep , it happens to me as well. I assume that Pawel is right that there is some number above which it reports a link instead of individual items
- Pedro Beltrao
It's from here: http://www.ncbi.nlm.nih.gov/bookshe... "If the number of citations retrieved is greater than the number of items displayed the feed will include a link to display the complete PubMed retrieval."
- Pawel Szczesny
Nice polymorphic API! They could also stick with one API and output more RSS entries in that case.
- Egon Willighagen
Possibly lame question. Is there any way to write ASN.1 format under BioPerl (Bio::SeqIO::entrezgene is a parser only)? Alternatively, is there a tool that converts from GenBank format to ASN.1?
I think there's a tool in the NCBI toolkit to convert between ASN.1 and other formats. Mentioned in this mailing list thread - http://bioperl.org/piperma....
- Neil Saunders
There is a binary and a text format for ASN.1. Writing the text format is straightforward. The ncbi toolkit contains an ASN1 compiler 'asntool' that reads an ASN1 format and produces the C code for reading the binary format. I tried to use it: cf. this old post: http://tinyurl.com/ybyq9x2 . Can you call those C functions from perl ?
- Pierre Lindenbaum
Neil, AFAIK NCBI's toolkit converts _from_ ASN.1 to other formats - it doesn't write it though. Similarly, EBI's toolkit doesn't support writing ASN.1. Thanks Pierre for the pointer, but since I need to write the text format only, probably it will take less time to put something together in Perl than in C. :)
- Pawel Szczesny
Ah right. Seems strange that everything is "read-only". Then again, ASN.1 is an ancient, internal format used and understood only by NCBI. They probably figured no-one in their right mind would want to write it :-)
- Neil Saunders
Unfortunately ASN.1 was abandoned in favor of XML (because it was missing some tools for non-C programs ? no tools for transformating too (xslt...) ) but IMHO, it is/was a very good way to structure the data: there is a schema to validate the data, the binary format is compact and the text format looks like and enhanced JSON...
- Pierre Lindenbaum
Pawel - the BioPerl module is based on a non-bioperl read parser Bio::ASN1::EntrezGene; have a look in CPAN. This doesn't go in the right direction for you either, but may be helpful. If ASN1 exists, something must write it! I'd ask the NCBI guys directly.
- Mark A Jensen
Also, try posting to the list : bioperl-l@lists.open-bio.org, someone may very well have done this before.
- Mark A Jensen
Just noticed that we used different notions of "public funding environments" in the mind map so far. What I had in mind was to have "funding environments" in public, much like what fundscience.org plan to do. Some of the added comments seem to have used the term in the sense of environments for "public funding". Both notions are certainly valid, and we should think of ways to keep them apart.
- Daniel Mietchen
good point re making this difference clear(er) in the map
- Claudia Koltzenburg
Yes, Jean-Claude, contests and prizes with a competitive element are definitely on the list. If you have good examples from the recent past, please post them here.
- Daniel Mietchen
"More money for science is always good. Or is it? Six experts tell Nature what concerns them most about the US stimulus spending and suggest ways to ensure that it benefits research and society in the long term." - http://www.nature.com/nature...
- Daniel Mietchen
Daniel - the thing I like about contests is that barrier to participation is orders of magnitude lower than traditional funding - there is no need to convince anyone that what you are attempting will actually work before doing anything. Of course this limits the type of projects that can be run but it still applies to a large number.
- Jean-Claude Bradley
Daniel, not that I have anything against HHMI, but that mantra is not exclusive to them. For example Max-Planck Society has exactly the same approach (and I would say that at 10% of HHMI's budget and having twice as much Nobel prize winners, MPG looks a bit more effective ;) ).
- Pawel Szczesny
Didn't mean this to be exclusive, and I am well aware of MPG approaches (been there for a while).
- Daniel Mietchen
"I wish there was a universal format for submitting grant proposals; authors could post proposals (once!) & then the funders bid on them." (rephrased from http://ff.im/5VwEI ). I would add that the process should be public. fundscience.org plan to go this way.
- Daniel Mietchen
How do funders and scientists rank "more attention to technological shifts" against the "scientific expertise they have"? One says "change" the other "keep doing what you know"! Are those two things not disagreeing each other? In other words, who would you fund first, the "crazy new idea" or the "conservative stuff"?
- joergkurtwegner
I would think funders should have (as they do now) the liberty of choosing their priorities, and in many cases this will be a mixture of many incremental projects and some revolutionary ones. The main shift in the system would thus be to have just ONE avenue for proposals, and to make it public.
- Daniel Mietchen
Reading anything right before going to bed is bad. Give yourself at least 30 minutes to unwind: lie on the sofa, watch mindless TV - anything that requires no brain!
- Neil Saunders
thx guys, were staying up late in the last couple of days due to thinking on coding problems hard instead of sleeping and realized that the iPhones next to my bed are not helping me, probably will move them to another room.
- Attila Csordas
Watching TV is a bad way to go to sleep. Reading books is great.
- Piaw Na
My 'going to sleep' material is usually science fiction or fantasy.
- Michael R. Bernstein
For unwinding I find that if you include a glass or two of red wine, - all of the above works :-)
- Nils Reinton
a good novel is my favorite way to relax before sleep
- Mike Chelen
another problem: I'm getting a Kindle tomorrow, can I use it in the bed to read fiction, comics, art... all the light things that makes us easy more or less?
- Attila Csordas
No art or comics, just fiction and non fiction.
- Piaw Na
Can you read the H+ magazine on your kindle ? Is it color ? Let me know if you can because I need to find a good way to read this and other pdfs.
- Nils Reinton
I could give you a list of selected publications that make me fall asleep whenever I try to read them ;-)
- Lars Juhl Jensen
I think the compatibility of IP and ONS is a very interesting topic - Steve Koch has an agreement with his research office that might make it partially compatible
- Jean-Claude Bradley
Jean-Claude, thanks for the pointer. I need to look into that. Also, Paulo Nuin commented on the blog that I could (and I will) mention data attribution.
- Pawel Szczesny
That's a great list. And data attribution and citation, maybe data management in general would be good additions. Paulo on the ball as always...
- Cameron Neylon
Agree with Jean-Claude that IP / ONS is interesting. Happy to provide any info on our very narrow and limited experience with the issue here at U. New Mexico.
- Steve Koch
Steve - hopefully we'll discuss that a bit at the ONS session in NC
- Jean-Claude Bradley
There're so many nice people out there - 16 hours after this posting (I finished the post on Friday evening) I got email with two useful publications in pdf format, because someone (thanks Piotr!) thought they would be relevant to the topics I'm studying.
- Pawel Szczesny
Watching now. BTW, you've convinced me to try "hundreds of slides" approach some day - I wasn't sure it works in more scientific talks.
- Pawel Szczesny
Pawel, I wasn't sure either, but I've had a chance to do deeper dives now (at Supercomputing) and it seems to work, especially as I get more comfortable talking science in this format. If I had to choose an alternate one it would be a pure storytelling/figure format, i.e. about some of your core results and work backwards if required.
- Deepak Singh
Heard at NGS Workshop that storage requirements for NGS are around 100-500 bytes per raw sequenced nucleotide and computing requirements scale accordingly.
- Pawel Szczesny
Yeh, just spread around the admin status amongst trusted people within groups and then delete/report when you see it.
- Cameron Neylon
The spammers don't think much of friendfeed, and less as time goes by. They mostly focus on facebook, but perhaps friendfeed might be seen as a facebook backdoor at some point.
- Mr. Gunn
I think, to be an OA challenge, you should not have phrased it as "list of publications on PubMed or" but "list of publications freely accessible via PubMedCentral or". Anyway, mine are not all biomedical, so I go for the "or" option: http://dbm.neuro.uni-jena.de/people... .
- Daniel Mietchen
NB deleted a comment because I don't know what Paulo is up to here, but I don't want to spoil it inadvertently.
- Bill Hooker
My only OA cred here is paying the OA fee for my dissertation, and pushing for that on an upcoming publication. Also doing a lot of repository work behind the scenes.
- Mr. Gunn
I'm afraid to get a virus on this link you just posted.
- Paulo Nuin
Ok I'll bite. Still working to get mine up on a single website but put a list in reverse chronological order on Paulo's blog. Basically of 12 papers since 2008, five are in OA journals, one is a CC-BY chapter in a book, and two others can be put online in final form six months after publication (which I haven't done in one case where I could). The ones that aren't proper OA I wasn't...
more...
- Cameron Neylon
"if Nature asks you for a piece you don't readily turn them down" - hard to disagree. At the same time, it's probably the best summary of the issues we are facing. Most of the discussion I had about future of science and impact of "2.0" meme onto academia were finished with a sentence along these lines :/
- Pawel Szczesny
Well there's two sides to it - one is the "Nature paper gives you credibility in grant proposals" element, the other is that it remains a more effective way of getting the message out than writing online or in OA journals.
- Cameron Neylon
Nature has been pretty accommodating - they agreed to make my book review OA http://www.nature.com/nature... - the other great thing about Nature is that they will publish work appearing on their open Nature Precedings
- Jean-Claude Bradley
That's true - I probably should have asked about that - didn't occur to me at the time...
- Cameron Neylon
There is prob nothing magic about 10k either... it's just the time that puts you into such a specialization, you shine because the competition does not. Something along the lines... in the land of the blind, one eye is king... or am I missing something important in the theory? Can the 10k be rewritten as function of gravity or so? Or perhaps the number of Higgs bosons per square second?
- Egon Willighagen
Pawel-- thank you for the link to the paper. Exactly the stuff that I was looking for.
- Kevin Gamble
Egon--point well taken. An issue with most things that are more complex like humans and all social science type research.
- Kevin Gamble
People don't realize that the server name is either unique or useless. Instead of giving you another example, here's a challenge: find a site related to bioinformatics named "project hope".
- Pawel Szczesny
My student Brian is trying to generate a 3D model of this protein - can anyone help? AEKYKAKKAKEKAYKKKAKEAKKAKYKAKEAKAYKAEKKAKYAKAKEKAYAKAKEAKAYAKAKAKAEKAKAKAKYAEKAKAAKYAEKAAKYAEAKAKAAEAKYAAEAKEAAKAAEAKYAAKAEAAKYAAEKAAEKYAKAEAAAEAKEAA.
SwissModel will give you a good "first pass", if there is a target with sufficient similarity. After that, he should learn Modeller - http://salilab.org/modeller/. Though now I look more closely - that's an odd sequence, unlikely to work using homology modelling.
- Neil Saunders
This looks like a coiled-coil protein and even more, something integral to a membrane (but I'm biased towards bacterial proteins, if this one is from Eukaryote, don't pay attention to my comment). If I'm right, judging from the length it should form four-helix bundle in the membrane. As, Neil, I don't think it can be modelled by homology at all.
- Pawel Szczesny
I too was going to suggest robetta - we've had good results but the queues are pretty long ...
- Anna Croft
Some similar responses in Portugal from people at the OA meeting. Big opportunities for countries outside of the big four (five? six?) to steal a march by working effectively together, particularly where there is focussed infrastructure investment.
- Cameron Neylon
Cameron, it's very interesting, I would like to know more. Is there any document out there or do you know some specific names of people I could get in touch with?
- Pawel Szczesny
from email
It was all very vague - I couldn't give you actual names because it was more just a positive response to the idea in general terms, but Portugal has something called the Knowledge Society Agency or similar, so that would probably be a good place to start http://www.infosociety.gov.pt/ Will try to put you in touch with them through the conference organizers.
- Cameron Neylon
Short commentary from my former colleague about MS Word correcting phrase "DNA polymerase" into "DANN polymerase" (without asking) in German medical textbooks and publications from German scientists. Unfortunately, not OA.
- Pawel Szczesny
Is this new? I had colleagues at Southampton working to exploit F based hydrogen bonding years ago.
- Cameron Neylon
Most likely not, but I bookmarked it, as I wanted to read something up to date on the topic.
- Pawel Szczesny
Ah ok - it just seemed like a recent paper claiming that it was all new. I could see that it might be the first comprehensive look at the structural effects.
- Cameron Neylon
What's the best way to embed a pdb crystal structure into a wikipedia article? Is there a gadget that allows reader to rotate & zoom? Or is static png the way to go?
I realized what I actually want: A widget that will display a molecular dynamics trajectory of the system, while allowing user to rotate, zoom, pull on atoms, etc. ... Comp. people get to work :)
- Steve Koch
One of the web widgets they offer at the PDB might work in the context of Wikipedia but I would guess this is one to have a wider conversation about? Andrew Su and Tony Williams would have an idea of the logistics at least.
- Cameron Neylon
The Jmol community has been trying for a long time now to get it available in Wikipedia... you could ask on jmol-users@ ...
- Egon Willighagen
As part of the Gene Wiki effort, we uploaded ~66k thumbnail images created by the EBI to wikicommons, and categorized them by SCOP. http://commons.wikimedia.org/wiki... Of course, they are auto-generated so they probably don't illustrate what you'd want to illustrate.. (and certainly not an MD trajectory...)
- Andrew Su
Thanks! Oh, and BTW, I was kidding about the MD trajectory, including the ability for the reader to pull on atoms in a real-time MD simulation. :)
- Steve Koch
Jmol has been up for discussion for a long time. I am not aware of any movement to support it at present. You might want to put the crystal structure hosted in Jmol elsewhere and make a link to it from the Wikipedia page rather than having it in the Wikipedia article itself.
- Antony Williams
Majority of software written few years ago isn't available for instant download. You need to print the license, sign it and fax it to the authors. License: 1,6MB pdf; software 180KB executable.
- Pawel Szczesny
I think the biggest problem is that often the scientists didn't even want to do that. In many cases it is their institutions that force them to put their software under a cumbersome license.
- Lars Juhl Jensen
I think those who received the invitation from Google itself got 20 invites to give (there is a wave with them). People invited by a regular user that had invitations can't invite others.
- Bruno C. Vellutini
"In fact, in the life sciences, the back and forth is getting more common, and will increasingly get so as the models for research, esp in large scale biology change. On the CS side there are any number of examples as well."
- Deepak Singh
To paraphrase Ken Robinson: the purpose of academia is to produce university professors.
- Pawel Szczesny
Short question: is there any up-to-date (let's say 3rd quarter of 2009) estimate how many scientific blogs is out there? If you have no idea, please give your own estimate. Biased consensus is better than nothing ;).
Thank you all. It's so little? I thought it's way more than 10k by now. It seems I wasn't offline for _that_ long ;)
- Pawel Szczesny
My guess is similar to those above -- maybe 2-3K blogs total, only about 500 active. But that's "blogs written by scientists or educators or librarians or other science professionals", not "blogs that sometimes mention something science-y"; I suspect the Technorati figure is about right for the latter.
- Bill Hooker