It is possible (or was, as I did it some time ago and don't have access to that installation anymore) using .loc files. See this guide http://wiki.g2.bx.psu.edu/Admin... - there's some info on Megablast customization. However I agree that Galaxy documentation is a bit lacking in administrative areas.
- Pawel Szczesny
For clustering (that is grouping by similarity) I use often cd-hit (http://cd-hit.org/ ). It has a special version designed to cluster fragments/reads from 454 sequencer: cd-hit-454. If more fine grained clustering is necessary, you can use CLANS (visual clustering software on the basis of BLAST similarity - http://bioinfoserver.rsbs.anu.edu.au/program... ). Finally, if the group of sequences is similar you can do MSA, as Larry suggested.
- Pawel Szczesny
Central thesis: "open access means that there will be more potentially harmful papers available to general public". We should get rid of the internet too, perhaps?
- Noel O'Boyle
Noel, that's not central thesis. That's example of reasoning to which Peter's approach leads.
- Pawel Szczesny
from iPhone
Pawel, I agree with your point, but I disagree with your example... allowing people to make mistakes I value more than believing you know better and disallow things... your argument is like disallowing freedoms because some cannot handle them... that makes your argument quite different from those of PMR, even though both anecdotal of shape...
- Egon Willighagen
Egon, I get the difference very well and that example was chosen on purpose (I had some less "nuanced" as well, but didn't decide to use them). However, if a scientist is using unscientific arguments, the remaining nuances don't make a difference anymore. How we can improve the quality of public discourse if we allow for such argumentation?
- Pawel Szczesny
Well, theoretical sciences I guess :) PMR formulated a hypothesis, that is worthy of empirical validation... which effect is larger: that of foolish people, of that of people kept uninformed... (I don't know; I'm not a social scientist...) Where would science be if we cannot hypothesize anymore.... You provided an important alternative hypothesis... null hypothesis, perhaps?
- Egon Willighagen
Ok - I get you now. Central thesis: "Anecdotal evidence is irrelevant to a scientific argument." And other things besides OA can help more.
- Noel O'Boyle
Egon, the problem I see is that PMR's own words: "I don’t think anyone can deny the truth of that conclusion." are not a formulation of hypothesis, but a populist language. This unfortunately has many implications, none of which I like.
- Pawel Szczesny
I agree that PMR's blog posts are often short on links to further detail. I personally prefer linking in my blog to further info to back up my story and improve the 'learnability'... I personally think 'learnability' is the more important aspect, and Open is the means. PMR is an established Cambridge scholar... they can do with bold language, leaving things to be worked out by others....
more...
- Egon Willighagen
Egon, you've touched a few very important issues. One thing is that if you don't have a leverage, you use the leverage of the community of which you're a member. So, PMR comments might influence position of people relying on the community in certain situation where trust in a discussion depends on the community's 'brand'. The other thing is that from perspective of people in less...
more...
- Pawel Szczesny
And finally, the biggest win of OA will be when the general public actually cares about open access to knowledge, not when scholarly communication will be open and still nobody cares. That will translate to _rational_ public discourse, because fact checking and hypothesis testing will be a normal mode of coming to conclusions on public issues. Indeed, there's a place for anecdotal evidence, but they serve as a starting points of coming to conclusions, not end points.
- Pawel Szczesny
In the past I have used SuMo http://sumo-pbil.ibcp.fr/cgi-bin... by Martin Jambon (also a BioStar user - Martin, where are you? :) ). It will require some modification of your PDB file so you can easily select which residues you want to take into account (for example by marking required residues as belonging to chain X, etc.). See the server for details. Alternatively, you could try to download one of many programs designed to analyze protein structures for binding pocket similarity (ten or so are cited in the introduction of this short paper: http://stat.fsu.edu/~jinfen... ) and then substitute program's database with your own definition of binding pocket. However, I can imagine it's not going to be the most straitforward way.
- Pawel Szczesny
My second impression (inspired by the plot on the screen) was that with proper software you could actually "play" the stock market on this stuff. :)
- Pawel Szczesny
What are the options available to parse output (full output with alignments) from jackhmmer? As far as I can test, BioPerl hmmer module isn't able to read it correctly. Are there any ready-to-use libraries (language doesn't matter), or I should write something on my own?
- Pawel Szczesny
Well, apparently the correct answer is that no library exists that parses full jackhmmer output. Its authors even write: The output format is designed to be human-readable, but is often so voluminous that reading it is impractical, and parsing it is a pain. The --tblout and --domtblout options save output in simple tabular formats that are concise and easier to parse. In case one needs sequences of hits, there are two options. One is to painfully write a parser. The other one is to extract sequence based on coordinates provided by tabular formats.
- Pawel Szczesny
Dear Open Science Friends -- I am going up for tenure this semester at the University of New Mexico, Dept. Physics. External letters are one of the most important components of my dossier. I am thinking that a co-signed letter from friends I've met in the Open Science community would be highly valued by those people evaluating my tenure case.
I am hoping that some of you would be willing to draft and co-sign a letter of support to be sent to my tenure committe--if so, I would be deeply grateful. Based on e-interactions with me, and in-person at ScienceOnline2010/2011 and ScienceCommons 2010 (Seattle), I think maybe some of you would have positive things to say about my research (I consider our lab's work in open notebook science, open data, etc. to be research) and service (promoting open science being external service).
- Steve Koch
The committee wants to see specifically: "..evaluation of the quality, importance, and impact of his research and scholarship, and, if possible, a comparison with other physicists at similar stages in their academic careers." and "...Any comments you wish to make based on your knowledge of his teaching, interaction with students, mentorship, and service will also be appreciated."
- Steve Koch
The standard method in our department is to mail letter requests to specific individuals, and to request hard copy signed letters. I've been thinking about this, and while I could ask one of you individually, it seems to me more appropriate, more powerful, and more fun to seek a co-signed letter. This has the risk of the various evaluation committees ignoring the letter, though. One...
more...
- Steve Koch
If anyone has better ideas, or knows of earlier examples of this from other universities, please post on here, or email me at stevekochscience@gmail.com
- Steve Koch
I am really grateful for the years of support I've received from many of you open science supporters in my early career. I know I've been super-lucky to even land the tenure-track job and have always known that. I know that tenure would be a huge gift that is not something I deserve or am entitled to. Instead, I look at it as part of the way things work now, and being awarded tenure...
more...
- Steve Koch
So what am I asking? Without better ideas for how to do this: I am asking for people in the "open science" community who feel they are familiar with some of my lab's open science or my teaching of open science to state a willingness to be a co-signer of a tenure support letter. Then I am asking that group of people to nominate one person to provide the official letterhead and sign (with...
more...
- Steve Koch
They are looking at October 1 for deadline. Thank you thank you all! And like I said suggestions for a better process are welcomed!
- Steve Koch
Do biologists qualify? If so, count me in.
- Pawel Szczesny
@Pawel, absolutely! And Thank you! Everyone counts--all fields of academia (people that come to mind immediately are biology, neuroscience, physics, chemistry, library science), industry, and even those who may not currently have a paid position, but are active open science supporters.
- Steve Koch
Count me in, natch. Regarding process: I think you are right to want individual letters from "high status" folks, to stay in the comfort zone of your committee. I suggest that the individual letters be different from the co-signed letter, which is a great idea and could be mentioned in the individual letters as "an instance of the wide web of collaborative opportunity opened up by Dr...
more...
- Bill Hooker
Do you have any ppt or whatever of your scio10 or scio11 presentations handy? Was there a recording? I'd be happy to do up a short letter and certainly to cosign a joint letter.
- John Dupuis
I'm in. I've served on tenure case and appointments-and-promotions committees, so I suspect a group-signed letter won't have as much weight as individual letters from as high up the academic feeding chain as you can go. Send me out-of-band email at gezelter@nd.edu if you want to discuss.
- Dan Gezelter
I tend to agree with Dan. I'm more than happy to sign a community letter, though.
- Egon Willighagen
I've never met you personally, so I don't think I should sign any individual letter, but I'd co-sign a letter of course! Just lt us know how you plan on doing that and I'll be there.
- Björn Brembs
I'd be happy to help, in any way. I'm so sorry to be a biologist, but at least of a physicist descent (my father), so maybe...? :)
- Marcin
I'll happily sign, not high up the feeding chain though I'm afraid :D
- science3point0
OK, here's a start: http://piratepad.net/DpuxZJT... I haven't written tenure case letters before so it may need considerable chopping about, but at least we can get started.
- Bill Hooker
Bill -- That is amazing! Thank you so much! That is wonderfully written and I appreciate it deeply. I've only been in a few departments in my young career, but I can tell you that UNM's physics department is full of faculty that care deeply about education and doing research for the benefit of society. So, I am confident that many of those voting on tenure will be very impressed by the...
more...
- Steve Koch
@Marcin I love interdisciplinary! Actually our science is focused on biology and have had students in the lab from biology, biomedical sciences, mechanical engineering, biomedical engineering, physics, and optical engineering. Moreover, I don't worry about the departments or academia / industry--with open science, we almost automatically ignore those credentials--while hard-earned,...
more...
- Steve Koch
@Dan Thank you for the experienced perspective! I've sent you an email with further thoughts / questions. I do accept the reality that I need those high-on-the-chain letters. And will seek those for sure. My gut is still telling me that at least for my department, a co-signed letter will add a lot of value, especially a letter such as Bill has drafted. I think for sure after the dust...
more...
- Steve Koch
@ John -- thank you for offering to co-sign! This is the only video I'm aware of (I think you were there & also my APOLOGIES for inadvertent slam on library journals!). That was the "data discoverability" session in 2011 and it was a ton of fun for me. Here is the link http://friendfeed.com/kochlab... and here are my notes from the session (with links) http://openwetware.org/wiki...
- Steve Koch
Here are my notes (with links) for the ScienceOnline2011 Data Discoverability session: and here are my notes from the session (with links) http://openwetware.org/wiki...
- Steve Koch
@others who've piped in--thank you! I am excited to have this support
- Steve Koch
Steve - I'm happy to co-sign but I do agree with Dan that individual letters are generally preferred. But if your tenure committee is saying that this is the way to go, definitely follow their direction.
- Jean-Claude Bradley
I'm happy to sign as well. Best of luck, Steve.
- Mr. Gunn
Now that people are back from science online, *bump*
- Bill Hooker
Thank you everyone! Jonathan Eisen has graciously signed and mailed in a snail mail copy of the open letter to my committee. I don't expect to hear anything about it (positive/negative/neutral) until December, maybe January. I will of course update you with any news. Thank you again, everyone: so much support made me feel great!
- Steve Koch
Pawel, what is your point? Should funders use a bad system because there is nothing better that allows them to judge some proposal in 5 minutes? (Because that's what these things are used for...)
- Egon Willighagen
Not that they should. They don't have any choice (article views is a joke - it can be gambled in no time).
- Pawel Szczesny
from iPhone
That's my point: they *have* choice, and the picked something bad. page views is not a better solution; I agree with that.
- Egon Willighagen
Egon, what is the other choice funders didn't pick?
- Pawel Szczesny
from iPhone
I quite like citations. It's +1 old-school style.
- Noel O'Boyle
Pawel, actually trying to understand what the contributions of a person to science are. Takes a bit more effort, but at least is fair, balanced, and transparent. No way to hide your decision behind vague metrics, but stand behind your decision. That's what funders should do.
- Egon Willighagen
Funders need feedback data to help get traction on understanding the contributions, I think. Citations (lists, rather than counts, particularly) have a role here, the dynamics and distribution of page views might, and lots of other metrics do too. Ideally would hand-curate a bit, then look at clusters, patterns, etc as part of the info that goes in decisions. You know, data mining for decision support, like grown up companies do.
- Heather Piwowar
i wonder if PMR knows about the MESUR project and all that. There are people trying to make sense and all out of usages. MJ Kurtz has looked into it quite a bit at ADS, too.
- Christina Pikas
@Christina yes I know MESUR. All these projects rely on grace-and-favour handouts from secondary publishers. I believe that what studies there are show that citations are considerably less accurate than usage in showing linkage etween disciplines, etc.
- peter murray-rust
Egon, I'm all for reading papers instead of measuring them. But that's what hiring commitees are usually doing, not funders. Funders have too many applications to use manual assessment on all proposals (Gates Foundation has a grant program in which they assess only the idea - author and his afiliation is hidden from the reviewers. The last time I was interested, announcement of the winners was postponed by 7 months, despite the fact that proposal was only 2 pages.).
- Pawel Szczesny
And that gets me to the original point - funders are interested in using robust and hard to gamble measures. No such alternative to counting citations exists (I was surprised to hear that funders are aware that if they start to _analyze_ citations, very soon researchers will use negative citations to 'downgrade' competition). Anything relying on clicks, views, usage, search engine results etc. is easy to gamble.
- Pawel Szczesny
That's bull. Citations can be faked too; it's harder, but happens too. Just write a review. Write something controversial. Write editorials. The thing here, this gambling is easier by those who are already on top, and have this currency to coin in some extra base level citation. But, I agree, if funders would *only* use citation counts for the papers, I'd be more than happy already....
more...
- Egon Willighagen
Egon, we all know that reviewing grant application is a mix of pseudomeasure of impact (whether it's a citation count, H-index or article-level metrics) and individual preferences of reviewers. I'm only saying that choice of pseudomeasurein practice isn't ours - it's funders'. Before NIH OA policy became mandate, comply rate was ca. 4% - now it's 75%.
- Pawel Szczesny
Are funding agencies and proposal evaluation discussed in Moneyball? I haven't seen the movie yet :) If agencies were rewarded for winning, really truly rewarded in a timely manner, then they'd figure out how to use citations, pageviews, everything available to predict impact of proposals. As far as I know we're a long way from this, since we can't even measure winning, really, for a funding agency, can we?
- Steve Koch
The rule of thumb is to postpone any genome-wide analysis until the genome paper is published. On the other hand, using deposited data to perform a vertical analysis (analysis of a single gene family across several species) is usually accepted. However, the best idea is to ask the sequencing lab. You'd be surprised how nice people are usually working there ;)
- Pawel Szczesny
Something outwith my control happened two days ago resulting in my display monitor appearing upside down. I've therefore had to physically turn my monitor upside down in the meantime. How do I change the settings on my PC (Windoze XP) to return things to normal?
Does ctrl+alt+up work (or ctrl+alt+down)?
- Pawel Szczesny
If the shortcut above doesn't work, try changing the registry key HKEY_LOCAL_MACHINE\SYSTEM\GDI\ROTATION "Angle"=dword:changehere to 0. Software to edit registry is called regedit (type in "run" box).
- Pawel Szczesny
TOP MAN, Pawel. ctrl+alt+up and/or down didn't initially do anything. Investigated option two but unable to find the sub directory I was looking for. Tried option one again and BINGO !! My life is no longer upside down.... Cheers !
- Graham Steel
If you count only papers you actually read, having a journal subscription makes individual papers much more expensive. $35? What about $50-$60 in a subscription? (numbers based on individual deals with large publishers of the institutions I worked at - "big deals" look quite different)
- Pawel Szczesny
Pawel, I don't understand your math... 50-60 is per paper? for the institute, for you personally?
- Egon Willighagen
If your institute pays for subscriptions of the whole journal, but you actually read only small portion of what's published, the cost per paper read is ca. $50-$60. Last time I checked it was true for a single institution and for a large society as well.
- Pawel Szczesny
It would be great to see statistics on that, although I guess these numbers are not widely shared.
- Marcus D. Hanwell
So, those are journals of which a whole organization reads about 100 papers, without depublication? With duplication that would go down to perhaps 25 papers, for one journal. Cancel, I'd say. You don't buy a newspaper just to read the front page either... Seriously, such a journal would fail the 'we think this journal has something interesting to say' bar, not? But I cannot imagine this to be the case for the average journal? How is this counted anyway? Like Marcus said, please, more detail.
- Egon Willighagen
These are almost always "back of a napkin" calculations, because deals with institutions or societes are usually negotiated individually, so it's hard to compare notes. Also, with subscription you don't only buy papers, but also convenience. That said, the number I've seen twice were quite similar - ca. 10-15% of available content actually being read by subscribers. Given the deals that were signed, that ratio elevated cost of a single paper to $50-$60 range.
- Pawel Szczesny
I guess that makes ~ $35 the best estimate. After all, the publishers have access to *all* needed information.
- Egon Willighagen
Yes, I'm sure all prices are not out of thin air, which makes me wonder if stats were the reason for publishers to sharply raise prices (MPG/Springer in 2007 or last year's UoC/NPG).
- Pawel Szczesny
That misses the point of the question, imo. It's a thought-experiment meant to help set goals, like asking "what would a perfect scholcomm system look like?". We can't build one of those either, but designing it on paper is a useful exercise.
- Bill Hooker
Bill, that was my reaction at first, too. But upon further reflection, I think that the point of the *answer* is that it's too easy to get lost in fantasies about what things should be like without really engaging with the real situation in real life. If I were trying to reform some large complex system that had many competing real-world interests, and people kept wantint to talk about the "perfect system," I suspect I'd lose patience with that line of thinking pretty quickly.
- Your Neighbor Steve
I wonder who will be the first to write a post entitled: "What's wrong with scholarly communication today? Wannabe stakeholders." ;)
- Pawel Szczesny
Steve++. Elements of the OA movement have a bad habit of bandar-loggist "Something noble and grand and good/Won by merely wishing we could."
- RepoRat
Strange, every time I click on "mentions" the phrase is right. Do you have an example?
- Pawel Szczesny
Oh, ic... maybe it's just the list where single words are highlighted... :)
- Egon Willighagen
Just checked: you're 100% right! the second col in the sum page are the account 'about me' bits, and there it highlights the single words :)
- Egon Willighagen
Institutional Repositories, Open Access, and Scholarly Communication: A Study of Conflicting Paradigms doi:10.1016/j.acalib.2011.07.002 http://www.sciencedirect.com/science...
Quote: "Figshare has been developed with zero cash (but a lot of love from Mark). That will scale as far as establishing that the concept works and scientists like it." Indeed, lots of love (and way more work) from Mark, but please Peter, get your facts straight. It's not that anybody else deserves credit for FigShare (nope - it's all Mark's work), but I think it's important to remember that zero cash approach _doesn't scale_.
- Pawel Szczesny
@Pawel I said "as far as". I believe that that point has already been reached. You may tell me different. I did NOT say it scales indefinitely - indeed I intended to imply that it would not scale automatically when it got beyond proof of concept. If you think it has not reached proof of concept say why and what it needs to achieve
- peter murray-rust
Thanks! I actually found a paper about identying EF-hand domains and I've already got idea what to do. But thanks for the link! It will be helpful.
- Mateusz Koryciński