The Life Scientists - FriendFeed

This is a public room. Anyone can join to post and comment.

Services

Administrators

Members

The Life Scientists

A room for all the life science types on FriendFeed (and everyone we've co-opted)

Neil Saunders posted a link

Genome Quilts by Beverly St. Clair

2 hours ago - via Bookmarklet - Link

Maxine, ﴾͡๏̯͡๏﴿ and Jan Aerts liked this

A different form of data visualisation - Neil Saunders via Bookmarklet

Donnie Berkholz posted a link

Mememoir: A Better Wiki For Science - ReadWriteWeb

3 hours ago - via Bookmarklet - Link

Deepak liked this

"Mememoir aims to provide a wiki that is heavily focused on authorship and can help to dispel the prejudices scientists have against publishing in a wiki-like format." - Donnie Berkholz via Bookmarklet

Donnie Berkholz posted a link

Important work can be done while daydreaming - The Boston Globe

3 hours ago - via Bookmarklet - Link

""The point is that it's not enough to just daydream," Schooler says. "Letting your mind drift off is the easy part. The hard part is maintaining enough awareness so that even when you start to daydream you can interrupt yourself and notice a creative insight."" - Donnie Berkholz via Bookmarklet

dekay posted a message

“Does anyone have a theory why exactly scientist are so "under"paid vs other disciplines?”

Wednesday at 11:34 pm - Link

Alexey and laura liked this

As Fox Mulder once said, I have plenty of theories :) - Neil Saunders

I love Fox Mulder. He had his priorities right - Deepak

Show 11 more comments

One theory: skills, experience, education are irrelevant. Contribution to "the economy" determines salary. It's hard to measure how scientists in academia contribute to GDP. Let's face it, many of them don't. - Neil Saunders

Because scientists are so much passionate that you can have them work even at very low salaries... so why pay them more? (basic supply and demand stuff...) - Enro

In addition to what Neil said, funding is wasted on the same experiments over and over again. It's not only about five groups working on the same problem today, but also about wasting resources on repeating experiments that were done 10 years ago. As far as I heard, money wasting is included in the science budgets, but that doesn't make measuring impact of science on GDP any easier. - Pawel Szczesny

It's not just scientists. Software Engineers (read MS in Computer Science) are pain $10-20K less that they might make if they were working in industry. They are doing the same kind of work AND getting appreciated much less. You have to really like scientific research to want to enable it in that kind of environment - Deepak

What Enro and Pawel said... - Björn Brembs

Supply and demand? Even now, with the relatively low salary, there are too many scientists (compared to the number of tenured positions). - Michael Kuhn

Pretty much just an echo here. Society doesn't value research appropriately because it's difficult to measure its value -- you need to think and look over long timescales, decades or longer, which people aren't used to doing and political cycles actively discourage our "leaders" (ha!) from doing. The NIH puts out a feeble little pamphlet "why fund science?" every so often, but I don't know of any concerted effort to make a strong business case for research. - Bill Hooker

There is a common belief that science is a noble profession. Society considers those who work in science to be privileged. The privilege conferred on those who can do science is supposed to compensate for lower salaries. - Jack H. Pincus

Jack, unfortunately, I don't see too much privilege given to scientists either these days (at least in the US). - Deepak

You're not in an income-generating field. You're also not generally employed to do something that management wants to get done so they can create a product: you're there because you've got the research bug. - Chris Cotsapas

Interesting to see how people are reading "scientists" as "scientists in academia" ;) Thanks to Chris & Bill for a different perspective… - dekay

It's because academic scientists are the ones that are the most underpaid, dekay. Industrial scientists are still underpaid relative to their education level, but nowhere near as bad as academic scientists. - Mr. Gunn

one of my theory that it happen because society & government can not estimate value (impact) of scientific products. It's not electronics or software or food... scientists produce papers!!! No doubt some of them will be translated and find application in medicine or other technologies but most of them not. So income (money that government or other foundations invest to academic science) much more then outcome - benefits for society and highly expected products. - Alexey

People are rewarded ('paid') based on how much 'obvious' (or relevant) contribution or product they make for their society. Scientists are not paid well because their contribution to the society is not really obvious. Their work can be presented in an important way (e.g. fight cancer) - and I think that's how scientists manage to get some money - but the problem is that for what they do, 99% of them do not generally have (or care) about an answer that is of beneficial to the society. Only 1% do. - JWS

Maxine posted a link

Evolutionists Flock To Darwin-Shaped Wall Stain | The Onion - America's Finest News Source

13 hours ago - via Bookmarklet - Link

Onion article, flagged by OWL (Dave Lull). - Maxine via Bookmarklet

Posted twice yesterday and today http://friendfeed.com/search?q... - PauloNuin

Thanks Paulo...once things drop off the first page of FF, it is too easy to miss them. I expect they are working on it. - Maxine

Cameron Neylon posted a link

Open Coffee - A(frican) Blog of Ecology - Raf Aerts' blog on Nature Network

20 hours ago - via Reshare - Link

Maxine liked this

Raf Aerts looks at taking an Open Notebook type approach to the study of coffee. - Cameron Neylon

dekay posted a link

Welcome to BioConductor — bioconductor.org

15 hours ago - via Bookmarklet - Link

I use rflowcyt with ggobi on a regular basis. I love it. The basic package has some dependencies that are a little tough to resolve in Linux. Installs in windows with no problems. Generally a big fan. - Zaki Manian

Chris Lasher posted a message

“Should I bold important descriptive statistics (e.g., total number of genes, percent of total with one condition) in the results section of my poster that are not included in a table? Is this distracting or helpful?”

Thursday at 9:13 am - Link

$0.02 from me: distracting. I'd prefer the section be structured so as to emphasize whatever you want me to take away from reading it, rather than be confronted with RANDOM SHOUTING in the middle of the text. :-) - Bill Hooker

I second Bill's opinion - Lars Juhl Jensen

Thanks guys. This was my intuition. KISS, right? - Chris Lasher

I am with Bill in this. Less text, less tables, more graphics. KISS - PauloNuin

I think you should bold them. The usability/readability of a poster is a lot like a website -- the vast majority of people quickly browse for something that catches their eye. See useit.com for more details. - Donnie Berkholz

supplementary data as handout? published on the web? ;) - dekay

dekay posted a link

“Does anyone know a site which does a good, and more detailed, review of "R" packages? i.e. say what they are used for? (R as in http://www.r-project.org/)”

Thursday at 4:18 am - via Reshare - Link

Pawel Szczesny, Chris Miller, Neil Saunders and Pierre liked this

R News? - Egon Willighagen

There's an R wiki; http://wiki.r-project.org/rwik...; don't know how comprehensive - Neil Saunders

Show 3 more comments

Documentation/examples is not a strong point of R, it must be said - Neil Saunders

R views might be what you're looking for: http://cran.at.r-project.org/w... - Michael Barton

This page is pretty good too: http://www.nettakeaway.com/tp/... - Michael Barton

Did you have a look at the views http://cran.r-project.org/web/... ? They contain a list of packages to be installed as well as a short description of what you can do, although those are rather task oriented. Otherwise the RSeek search engine (http://www.rseek.org/) has become my favorite source of information. - Yann Abraham

@all: thank you for your links: here is my summary: http://www.twine.com/item/11fz... - dekay

Visualization packages: http://addictedtor.free.fr/gra... - dekay

Wubin Qu posted a link

BioPythonGUI

19 hours ago - Link

imabonehead and Anders Norgaard liked this

BioPythonGUI is a Graphical User Interface of BioPython. BioPython with GUI can make everyone use BioPython not only the programmers. - Wubin Qu

Deepak posted a link

Semantic Provenance for eScience: Managing the Del

17 hours ago - via Bookmarklet - Link

"Provenance information in eScience is metadata that’s critical to effectively manage the exponentially increasing volumes of scientific data from industrial-scale experiment protocols. Semantic provenance, based on domain-specific provenance ontologies, lets software applications unambiguously interpret data in the correct context. The semantic provenance framework for eScience data comprises expressive provenance information and domain-specific provenance ontologies and applies this information to data management. The authors’ “two degrees of separation” approach advocates the creation of high-quality provenance information using specialized services. In contrast to workflow engines generating provenance information as a core functionality, the specialized provenance services are integrated into a scientific workflow on demand. This article describes an implementation of the semantic provenance framework for glycoproteomics." - Deepak via Bookmarklet

Graham Steel posted a link

Do 68 Molecules Hold The Key To Understanding Disease? - Science Daily article

19 hours ago - Link

Eric Jain posted a link

Curating the Dark Data in the Long Tail of Science

Friday at 6:03 pm - via Reshare - Link

Dealing effectively with "long tail" data is an important challenge, great to see that some people are giving this some thought... - Eric Jain

Andrew Su posted a link

Wikipedia:WikiProject AP Biology 2008

Friday at 4:05 pm - Link

Martin Fenner, Maxine, Sung W. Lim, Kevin D. White and Deepak liked this

... at the risk of seeming wiki-obsessed, I was impressed by how enthusiastically some of those students are taking to the project. - Andrew Su

wiki-obsessed ... never :) - Deepak

Andrew Su posted a link

Broad is targeting a $1B endowment...

Thursday at 10:21 am - Link

Chris Cotsapas and Neil Saunders liked this

Wow - Deepak

A billion dollars ain't what it used to be: http://en.wikipedia.org/wiki/L... - Chris Miller

Show 6 more comments

True, but the Broad is a research institute devoted to biomedical sciences, not a general education university. On top of that, they've only been around (officially) since 2004. That's growing at a pretty good clip. (Nice when money comes in $400M chunks...) - Andrew Su

I presume that was the big news we were all waiting for? - Deepak

and IMO the future of a lot of science lies in places like the Broad, i.e. more innovation in institutes and non-profits than in university research labs - Deepak

where are the Australian billionaire philanthropists... - Neil Saunders

not in Australia, and not very philanthropic :P - Andrew Perry

:) yeah, who is there apart from Packers and Murdochs - Neil Saunders

Big data, big money, big scientists? - Attila Csordas

Attila, it would be great if such combination always mean "big science" ;) - Pawel Szczesny

The key is not big money, but not being in a place where feel the need to game the funding system, or write grants just beccause you want some funding - Deepak

Chris Lasher posted a link

My poster for the VBI 2008 Symposium, giving a brief idea of what I work on.

Friday at 12:43 pm - Link

Sally Church liked this

laura posted a link

Seeing the Signaling Forest and the Trees -- Yaffe 1 (35): eg5 -- Science Signaling

Friday at 12:28 pm - via Bookmarklet - Link

hubfeed posted a link

Auto/manual categorisation of PubMed articles

Friday at 8:44 am - via Bookmarklet - Link

Thomas Lemberger liked this

"Compare: * MeSH terms that are applied manually (with assistance from UMLSKS) to a paper by PubMed. * Wikipedia terms detected in the abstract by GoPubMed. * Categories applied to those pages in Wikipedia." - hubfeed via Bookmarklet

Mitchell J Stanton-Cook posted a message

“Feeling depressed as writing a document to become a confirmed PhD student. Keep thinking "how much is a PhD student really worth?" Money is broght in by work that leads to grants for the PI, money from publications... I'm beginning to think the AU20K is really harsh when you can make that on welfare”

Tuesday at 7:40 pm - Link

Alexey and dekay liked this

if you think about it though, it's not just work that you're providing but also training/education that you're receiving. in addition, it might filter out those really interested in science.. (don't expect me to be this understanding of the powers that be once i start next month ;) - laura

Just make sure you enjoy what you are doing. Money isn't everything after all. - imabonehead

Show 15 more comments

I'm not a money person (if I was I would be doing something I'm not passionate about) but it would be nice to be paid on the correct side of the poverty line. I just feel a little bit ripped off by the government with their lack of respect and understanding. Rent doubled in 5 years, petrol prices almost doubled in the same time frame, food prices.... - Mitchell J Stanton-Cook

I've been told today by a PI that science is worlds second oldest profession. And IMHO it _IS_ better than the oldest one;) - marcin

$AU20k does sound a bit low to me (I assume that's Australian). It was $14k in my day and that is going back a bit now. But I think it is important to remember its a stipend, not a salary. And that you are also paying the penalty for the proportion of students who don't make a contribution and suck up resources. I guess the stipend is still set by government though in Oz or is there a market? When I left it looked like there was going to be a bidding war for the best students. - Cameron Neylon

There's always Poland where you could get $5000 a year and gas is $2.4 per liter (ca. $9-10 per gallon, all are US dollars). :) But little more seriously (but not really serious), you should start getting used to feeling of being underpaid. I'm yet to meet a scientists that is getting paid what he/she is worth :) - Pawel Szczesny

Cameron: There is a thing called an APA-I ... where an industry partner contributes some "top-up" money (usually just AU $5K/pa). There doesn't seem to be many APA-I scholarships negotiated in the biosciences (at least where I am), although engineering students seem to get them. I think PIs / students avoid them because the hassle of the IP issues (including sometimes 'sealed' thesis chapters) is not worth $5K/pa. Maybe if it was easier to quantify the value we generate .... - Andrew Perry

Compare it to your undergrad times: education isn't free. And this time you are getting paid for learning. Just make sure you really like what you are doing, because you are not in it for the money. Never will be. - dekay

Agree with Laura - it's a prequalifier for serious scientists. Yes is an expensive choice, like all education, but doing a PhD 20 years ago really helped my career in the pharma industry and it was well paid compared with doing a Post Doc. I decided to take the pain early for the gain later. - Sally Church

I didn't mind my PhD fellowship at all. First time I had ever earned anything, so it wasn't that bad, plus I didn't have to pay any tuition fees. As others have pointed it, it makes sure you're committed. Now once you have a PhD and get paid a pittance, that I have a serious problem with. - Deepak

What Deepak said. A few years of frugal living is no big deal, it's when you come out the other end of the meatgrinder and your real, after-tax income goes *down* instead of up that's a problem. Bear in mind also that the opportunity cost of a PhD is unrecoverable: you'll never catch up the lost earnings even if you calculate them at minimum wage. Only do this if you really, really have a jones for research, because otherwise it Will Not Be Worth It. - Bill Hooker

Also, Sally and Laura, I strenously object to the idea that you can only be a "serious scientist" if you have a PhD. I doubt that's exactly what you meant, but it's a common and absurd and offensive trope. A PhD is an endurance test, nothing more. It's not especially good training most of the time -- very dependent on supervisor, most suck -- and it's certainly not any kind of quality control, just look at the many morons who get PhDs by some witches' brew of politics, asskissery and blind perseverence. - Bill Hooker

@Bill: I absolutely didn't mean to say you need a PhD to be a 'serious scientist'. Doing a PhD for little money when one could be making big bucks in some other sector might, just might, show some commitment to science (or let's say it should...). Not doing a PhD does not imply lack of commitment or interest. Regarding the quality of training and being happy with what little stipend you may get, I prefer sticking to my idealistic assumptions for now, as I'll be starting next week. - laura

@laura: I say (and believe) that "most suck" but that doesn't mean they all do, and in particular if you are aware in advance of how the system works (and fails) then you are in a good position to make it work for you. Just the fact that you are on FF means you are *vastly* better informed than, say, I was when I started my PhD. So let me try to offset my negative tone above and say: welcome to the community, good luck, and let me know if there's any way I can help you. - Bill Hooker

Bill, my apologies. You're right, didn't mean it in that way at all, more as an indication of commitment. Like Deepak, I enjoyed mine, worked hard and didn't mind 3 years of frugal living in London. We had a lot of fun and good supervisors/facilities, so it can be done. - Sally Church

Only three years.... I always forget that it's so short in the UK. I just started year seven of my PhD this week. I'll be done in a few months, but *still*. And my lab mate is in her 8th year, and another one did a MSc in 5 years. There need to be better rules in North America about when you can *stop*. - Eva

it is rarely actually three years - the UK system does encourage finishing but it does so by creating a situation where you spend 6-9 months working with no money whatsoever. The US situation there is usually some money but no sense of a strict deadline as I understand it - is that fair? - Cameron Neylon

I think in general, we could do better at teaching PhD students time management. The basic strategy seems to be: thrash wildly for X years, then write up whatever you have. Start with 3 months of planning with a timeline at the end of it - even if you don't follow it to the letter. - Neil Saunders

I think at UQ they are changing this now. There is no need to write up a thesis. graduation depends on papers not how well you write your magnum opus which if you're lucky more than your supervisor and the reviewers will read. - Colin Archer

Don't know what it's like in Aus now but when I went through ('98) the average for a full-time PhD was close to 5 years -- and the stipends were 3 years with a hard-to-get 6 month extension. I finished my final 18-24 months (took nearly 5.5 yrs) on $12K that I borrowed from my parents (v lucky to have that opportunity, too). - Bill Hooker

Neil Saunders posted a link

Big data: Wikiomics : Nature News

Wednesday at 10:10 pm - via Bookmarklet - Link

Alexey, Todd Harris, Konrad Förstner, Pedro Beltrao and 2 other people liked this

"But will the bench scientists participate? "This business of trying to capture data from the community has been around ever since there have been biological databases," says Ewan Birney of the European Bioinformatics Institute in Hinxton, UK. And the efforts always seem to fizzle out. Founders enthusiastically put up a lot of information on the site, but the 'community' — either too busy or too secretive to cooperate — never materializes. So how do the wiki proponents know that this time around will be different?" - Neil Saunders via Bookmarklet

IMHO, many previous efforts have been too focused on properly structured/formatted/annotated data to facilitate downstream mining. New emphasis (e.g., with wikis) is on lowering the activation barrier to participation. We bioinformaticians can worry about structuring that data later -- asking users to do it is a recipe for nonparticipation. - Andrew Su

Show 3 more comments

Great summary of the potential and problems of using wikis in biology. One issue is credit-assignment. WikiGenes links every piece of text directly to its author. - Martin Fenner

@Andrew: Regarding "activation barrier", you can't go much lower than having a simple text feedback box where people can post additions or corrections to existing data (without having to sign up or learn any special syntax or rules). So for databases that want to increase the amount of feedback they get, it may make more sense to focus on making that mechanism more attractive first. Not quite as sexy and grant-attracting as building yet another wiki, of course... - Eric Jain

If people care about the data, and it brings value to them, I believe they will participate. I am completely with Eric though. Some wiki's (and other services) make it difficult enough to sign up that people just can't be bothered. - Deepak

Eric, I think we agree... Didn't mean to suggest that wikis are the end all and be all. Blogging, twitter, even FF's "like" button are all in the same vein. Oh, "instant feedback" I think is another property that promotes participation. Surely every database has a comments/corrections email address that people *could* use to submit feedback. But those lack instant feedback, which is why I think they're underutilized... - Andrew Su

Andrew - On the other hand, if people can't rely on the fidelity of the data, will they actually use it? Email -> curation -> display gives users confidence in the data. And what is the critical mass required for community annotation? The more users, the more likely others will join. How many will also monitor and correct data submissions? How many are necessary? - Todd Harris

Todd, I think users are sophisticated enough to understand data at varying degrees of reliability. Sure, if you munge all your high-quality curated data in with instant contributions, that won't inspire confidence. But if the annotation model is clear for each section of your site, I think people will be fine with that. Plus, these user-contributed factoids then can become the low-lying fruit for the official curators (rather than staying on top of Pubmed, which is impossible). - Andrew Su

Andrew Perry posted a message

“Job Ad: Grimwade Research Fellow (5 year term), Department of Biochemistry and Molecular Biology, University of Melbourne.”

Thursday at 5:05 pm - Link

Neil Saunders, PauloNuin and Pedro Beltrao liked this

This position has just been advertised in the Department I work in. It's a pretty good deal for any senior Postdoc ready to make the transition to PI. - Andrew Perry

Andrew, how do you think your dept would view Open Science? - Bill Hooker

My feeling is that the Department wouldn't have any problem with Open Science per se, however the usual currency of high impact peer reviewed publications and acquiring grants still would apply, like most places. Those who are publishing well and bringing in grants to fund their research tend to be more highly regarded and do less teaching ... I can't imagine the Department really cares either way about the 'details' of how this type of success is achieved (Open vs. the status quo). - Andrew Perry

Thinking of applying Bill? - Neil Saunders

Sorta... it's a bit out of my league I think, plus I did just spend 6 years and tens upon tens of thousands of dollars to get the US to allow me to stay here... I dunno. - Bill Hooker

Chris Lasher posted a message

“What's an effective way to share a poster on FF?”

Thursday at 11:30 am - Link

I think you should upload it to Flickr in great resolution and then use the Share on Friendfeed feature with which you can share both the link and the image with us. - Berci Mesko

Upload it anywhere, and post the link on FF? - Eric Jain

Flickr, website, you name it - PauloNuin

Post to Nature Precedings and provide link? - Maxine

Didn't we just have a conversation about having conversations again...? Some more ideas at http://friendfeed.com/e/4e0946... - Cameron Neylon

Thomas Lemberger posted a link

PubMed search in Ubiquity

September 2 at 12:18 am - Link

laura, Rajarshi Guha, PauloNuin, Deepak and 12 other people liked this

I am not a programmer and have zero idea about Javascript, so had many problems. Improvements most welcome! Couldn't use properly the xml from ncbi's E-utilities (wrong mime-type???) :-( Anyway, to use the command, copy-paste the code into Ubiquity's command-editor (read more: <http://is.gd/1Wns>). Try a PubMed search by selecting something on the screen, call Ubiquity (option-space) and type 'pubmed', wait and see... - Thomas Lemberger

nice examples on the Ubiquity homepage - Pedro Beltrao

Show 4 more comments

Brilliant, I love it.. Didn't seem to save the command when I restarted the browser though - Jason Winget

added a little feature to select number of records displayed (eg include #3 somewhere in the query to limit to 3 records) and moved the code to http://code.google.com/p/tl-ub... - Thomas Lemberger

For parsing XML, jQuery is very nice and it's included in Ubiquity. No need to do match() :) - Rajarshi Guha

mmh, I tried this but it did not work... unfortunately ncbi's e-utils return text/html but jQuery expects text/xml (<http://is.gd/bu6>). Is there a workaround? - Thomas Lemberger

if retmode is xml, I can work with it in jQuery. Also in jQuery.ajax() you should set the type as xml. Also, namespaces need to be separated from the element by ':\\'. See http://groups.google.com/group... - Rajarshi Guha

I have a PubMed/JSON proxy in Scintilla if you need it - would make this much easier. - hubfeed

var parser = new window.DOMParser() - Thomas Lemberger

Bill Hooker posted a link

twidox: a free, user generated online library... publish, distribute, share, and discover

Wednesday at 9:18 pm - Link

"Documents on twidox are accessible to everyone online and will allow people to share their knowledge and help others with their work, learning, teaching and research. The focus of the website is on: * professional and industry specific documents * research material * academic papers and articles * coursework and dissertations * data and statistics" - Bill Hooker

Currently in "private beta"; might be a useful preprint/presentation server. via Peter Suber http://www.earlham.edu/~peters... - Bill Hooker

I'm not quite sure what this is supposed to be... - Björn Brembs

I signed up for the "private beta" but have yet to get my account info. I'll post more when I have a chance to tinker with it. - Bill Hooker

Neil Saunders posted a link

Big data: The future of biocuration : Article : Nature

Wednesday at 9:46 pm - via Bookmarklet - Link

laura, Alexey, Bill Hooker, John Dupuis and 9 other people liked this

"Third, curators, researchers, academic institutions and funding agencies should, in the next ten years, increase the visibility and support of scientific curation as a professional career" - Neil Saunders via Bookmarklet

This article by far the best in the special Nature issue this week. - Neil Saunders

Show 7 more comments

It's not a free access issue is it? - Deepak

Not sure Deepak - it sure should be, if it isn't. - Neil Saunders

I think the articles are freely-available; follow the links from http://www.nature.com/news/spe..., they are not marked as pay/subscribe - Neil Saunders

The entire issue is free online, but for two weeks only. Download them now while you can... - Andrew Su

exactly Andrew - Graham Steel

Downloading furiously - Deepak

/me is safe in the knowledge the dead tree copy will be waiting for him when he gets home :) - Daniel Swan

Yes it is free access for 2 weeks. I've posted twice about that on this FF group, see here: http://friendfeed.com/e/1489a2... (at time of writing, about 3 or 4 entries below this one). - Maxine

"The exponential growth in the amount of biological data means that revolutionary measures are needed for data management, analysis and accessibility." -- Is the revolutionary measure free access for 2 weeks......? - Frank

I can't wait to read Cory's article!!!! I am thrilled!!! - PauloNuin

Maxine posted a link

Event on 24 Sept: Scientific Researchers and Web 2.0: Social NotWorking? from Scientific Researchers and Web 2.0: Social Not Working? forum on Nature Network...

Thursday at 7:58 am - via Reshare - Link

"Second of British Library "Talk Science" quarterly cafe scientifique evenings. Timo Hannay talking. London, 6-8.30 p.m. I'll be there. - You via Bookmarklet (edit | delete)" - Maxine

arfon posted a message

“Just getting ready for Matt's talk: "Genomes on Rails"”

Thursday at 1:32 am - Link

Adam Kraut, Michael Nielsen, Sutee Dee, Sally Church and 3 other people liked this

Matt starts with "previously". How was sequencing done in the past: 15 years, $3 billion. Most of the bioinformatics work was done with Perl. - Martin Fenner

Onwards: compare genomes: species and individuals. Much cheaper and faster, powered mostly by Ruby on Rails. - Martin Fenner

Show 21 more comments

1 genome, 10 years. Upcoming....1000 genome project - arfon

2 main problems facing modern Biology: 1. Coping with rapid change - arfon

Data is rapidly changing, created virtual fields in their databases to handle this. - Martin Fenner

Virtual fields can be edited on the fly to reflect changes in the lab process - arfon

Problem 2. Throughput!!! - arfon

Ruby and Rails is also used to definy the workflow with a web-based interface. - Martin Fenner

Issues with large data sets: 75 TByte per week. - Martin Fenner

75 TB per week - arfon

Maintaining performance: denormalize the data - arfon

Large data sets are denormalized for faster queries. Uses multiple applications for scaling. - Martin Fenner

Run multiple instances of the same application for difference science domains - arfon

Deployment details: Nginx reverse proxied to a pack on mongrels - arfon

Automate everything. - Martin Fenner

Integrate with legacy databases. Uses RESTful services. - Martin Fenner

Trillionics: scale changes the way we do science. - Martin Fenner

Towards a virtual institute - arfon

Lots of data, lots of people, lots of compute, lots and lots and lots - arfon

Towards a Virtual Institute. Lots of data, lots of people, lots of compute, lots of uses. Issues of distribution. - Martin Fenner

Talk over. Questions: Have we considered anything other than relational databases? - arfon

Answer: not yet! Certainly will be looking at alternatives in the future - arfon

Thanks for Arf and Martin for the live updates! Really enjoyed giving this talk. Good questions afterwards and conversations over coffee. Excellent. - Matt Wood

This was the only science-related talk at this RailsConf. Most people seem to use Ruby on Rails for other stuff. Interestingly, the talk coincided with the special feature (big data) from today's Nature. - Martin Fenner

@Martin - a miracle of scheduling! - Matt Wood

One day I'll get to see you give one of these talks in person - Deepak

Maxine posted a message

“A reminder about the “Big Data” special in this week’s Nature. The articles in the package look at how massive influxes of data are changing the way science is done in many fields, and includes a feature story on ‘Wikiomics’ (written by Mitch Waldrop). FREE for 2 weeks from today 4 Sept.”

Thursday at 2:37 am - Link

Heather, Hilary, Michael Nielsen, Deepak and 5 other people liked this

Continuing the message above: all the articles, as well as downloadable PDFs of the print versions, will be free online for two weeks at http://www.nature.com/news/spe.... Please download everything you are interested in—and spread the word to friends and colleagues about what you like (and don’t like!) via email, blog, and Web site. And of course, don’t forget that Nature always welcomes letters to the editor at corres@nature.com. (Posted on behalf of Mitch by Maxine.) - Maxine

Thanks for the tip - ivanandersson

The Petacenter article on CERN is excellent but I'm getting a corrupt file message for the PDF - Sally Church

Neil Saunders posted a link

Moving Online at Bench Press

Wednesday at 10:05 pm - via Bookmarklet - Link

hubfeed, Bill Hooker, Deepak, Heather and 2 other people liked this

"I’ve used both paper and electronic notebooks, and I’ve come to one conclusion: whatever you can make electronic, make it so. Trust me, it’s worth the effort" - Neil Saunders via Bookmarklet

Quote from the post, not me, obviously :) - Neil Saunders

I really really like this. The more people who test and try out different tools, the more it will be possible to distill a consensus opinion on generically versus specifically useful functions. - Heather

arfon posted a message

“Genomes on Rails: '..previously at the Sanger institute 4500 Perl modules!'”

Thursday at 1:49 am - Link