Khader Shameer › Likes

Justin H. Johnson
Alignment-free estimation of nucleotide diversity.
Personal genomics, single-cell genomics =>Whole-genome molecular haplotyping of single cells with a microfluidic device
myKaryoView v2 Released: Navigate Your Own Genome!
Nir London
Jeff Habig
Ome of the season - "Carolome: Functional Imprints of Culture Memes in the Global Genome" A must read for the holidays!
The probability of a meme occurring by chance was calculated using the Assange principle (Assange 2010). - Jeff Habig
The meme FALALAA is also easy to spell even for a unicellular prokaryote. - Jeff Habig
The second most common meme in the Carolome is KISKLAS corresponding to ‘I Saw Mommy Kissing Santa Claes’. The University of NorthPole endorses The Ronettes version from 1963. This is the most dance-friendly version recorded (Palin 2010). - Jeff Habig
Sally Church
Has anyone tried the 23andme genome test at all? I finally caved as it dropped from $499 to $99, but somewhat sceptical about what the results might be. Curious to see if there are any impressions from the group here.
For $99, I may join you in finding out! - Mickey Schafer
cool -- thanks, Neil. - Mickey Schafer
For me it was more of an interesting geeky sort of gift, like buying a gadget. It was interesting in general but not much more than that. - Pedro Beltrao
Fun and also Useful to us. Found a sizeable genetic predisposition to something parents experiencing. Prevention path includes vitamins. Starting now rather than 20+ years from now when symptoms start to turn up. - Heather Piwowar
Was also contacted by genealogical group related to my background (, and as a result have learned the names of ancestors from 1600s. Fun :) - Heather Piwowar
Interesting offer, however it requires one year commitment to the 23andme's Personal Genome Service subscription (at least 60 USD). With intl. shipping costs (to Europe) it makes than almost 260 USD in total. - hlbiotech
My wife and I just ordered the kits last week. No more experience yet... - Antony Williams
Carl Boettiger
How do you manage citations when writing on the web (blogs, etc?) Would be nice to have a reference manger that works in browser editors like #Mendeley does in office docs
It would be really nice if the Google Docs API supported the right hooks for use to do this. - Mr. Gunn
When you say manage, do you mean insert during the writing process? - AJCann
Yup, that would probably be the most valuable. For instance, even a Wordpress plugin that let me cite using the bibtex reference would be quite nice, if such a thing exists? - Carl Boettiger
Good news: bibtex wordpress plugin exists. bad news: formatting fails on my site. - Carl Boettiger
Got a basic citation system working in wordpress at least: - Carl Boettiger
My thoughts on this are here: Briefly, we should integrate citations as Wordpress links, this would give us a lot of interesting features for free. - Martin Fenner
Thanks Martin, that's an excellent piece. Yes, I agree citations should be put in as links, though I'm not sure what links: as @rmpage points out, doi based links might make more sense than those pointing to a particular publisher? What track-back tools are available that could determine how many incoming links a paper has? How do these cope with multiple potential link urls to the same paper? - Carl Boettiger
Carl, you point a big problem that goes beyond managing citations for a blog. I'm a big fan of DOIs, and I'm always confused why so many places prefer their own internal identifier (PubMed is a prime example). - Martin Fenner
Next generation of algorithms inspired by problem-solving ants | -
Next generation of algorithms inspired by problem-solving ants |
"An ant colony is the last place you'd expect to find a maths whiz, but University of Sydney researchers have shown that the humble ant is capable of solving difficult mathematical problems. (...) To solve these optimisation problems using software, computer scientists have often sought inspiration from ant colonies in nature - creating algorithms that simulate the behaviour of ants who find the most efficient routes from their nests to food sources by following each other's volatile pheromone trails. The most widely used of these ant-inspired algorithms is known as Ant Colony Optimisation (ACO). (...) "Even simple mass-recruiting ants have much more complex and labile problem solving skills than we ever thought. Contrary to previous belief, the pheromone system of ants does not mean they get stuck in a particular path and can't adapt. Having at least two separate pheromones gives them much more flexibility and helps them to find good solutions in a changing environment. Discovering... more... - Amira from Bookmarklet
See also: Ant Colony Optimisation - Amira
Wow... this was news when I was PhD... like, 8 years ago - Egon Willighagen
well.. nobody said that ACO algorithms are the latest news ;) but Reid & Sumpter's tests using the Towers of Hanoi puzzle under dynamic changes are.. Here is a research article - Amira
Iddo Friedberg
Most information is embargoed, so no microblogging. Will write a blog post soon. - Iddo Friedberg
I'm assessing the predictions for the CBS dataset. Interesting exercise: - Iddo Friedberg
Steven Brenner about the embargo rules: do not talk about other people's prediction's by name. - Iddo Friedberg
I just gave my talk on the assessment of predictions of the CBS dataset. Wow, I really did put in a lot of work. - Iddo Friedberg
Björn Brembs
Björn Brembs
Shirley Wu
Get it while it’s hot! 23andMe for $99 -
cutting to the chase: use discount code UA3XJH ( - Shirley Wu
Nice. Found out in time, this time round - but why don't they ship to Maryland? - Rajarshi Guha
@Rajarshi: Certain states do no longer allow this kind of testing. Just ordered mine. - Josef Scheiber
Just tried that code at that link and it is coming up as invalid - it should still be working on Nov 26, right? - Jean-Claude Bradley
I went on tech support and got it resolved with the discount even without the discount code - looking forward to the results :) - Jean-Claude Bradley
Shirley Wu
The perfect gift for the genetics geek, genealogy buff, or science nerd that you know (or maybe just for yourself ;) - get 23andMe for $99 today and tomorrow using this discount code ($400 off regular price): UA3XJH (
Whats with the new $5 per month subscription ? :) I was going to buy a bunch for xmas gift but this might put me off a bit. - Pedro Beltrao
Dropping the base price to $99 requires us to collect additional revenue, unfortunately. Think of it as a $159 gift? with option to continue? We continuously update the content so there it's not a one-time value. - Shirley Wu
Can that be canceled any time? - Rachel Walden
@Shirley - sure, it is still a great gift. - Pedro Beltrao
I think I may buy this as a christmas present for someone. I will be sure to point them towards the store page showing the full price. - Michael Barton
@Rachel, there is a 1 year minimum at purchase but it can be canceled anytime thereafter - Shirley Wu
Thanks, Shirley - Rachel Walden
Thanks Shirley! Just ordered mine. - Rob Syme
Ordered mine, blogged and tweeted about it, and now pondering whether a Barcamp-style meeting makes sense once Genomes Unzipped allow data contribution. Sigh :-) - Oliver Hofmann
also add ~$90 to the cost in case of international shipping - Attila Csordas
New science: The puzzle of biological diversity
Andrew Su
Survey: How many scientific articles do you estimate that you read per year? (really read, not just skim...)
Maybe a couple of dozen (not including reviews -- I probably do 20-25 of those a year, and each requires at least one close reading). I probably skim 5 or 10 times as many as I really read, where "skim" = glance through intro/discussion and take a reasonably close look at the results, usually to pull out some particular piece of evidence for my own use. - Bill Hooker
3000/500/100 (read abstract and saved for later/skimmed/read carefuly) - Pawel Szczesny from iPhone
(Remember, Pawel is The Man Who Reads Everything. Mere mortals should not be judged by this standard.) :-) - Bill Hooker
10-15 across all fields these days. When I was still in the field probably half a Pawel - Deepak Singh
Really read in-depth? Probably a dozen. For most articles, I just skim and take the most relevant parts. - Chris Miller
carefully read: 200 (I have time-slots devoted solely to reading ~10h/week), re-do the math: 20 - marcin
From 1997 to 2007 I saved 1350 PDFs on my hard drive. So 135/year. I think I read all of them fairly carefully. In the 20th century I also read actual paper copies so more, maybe about 200. Now I read less and skim more, no paper copies at all, lots of abstracts by RSS feed. - Dave Lunt
Read carefully - several dozen per year. Skim/read selectively - hundreds. - Mickey Kosloff
I'd say I was around 1000/500/100 when I was in grad school - Mr. Gunn
@MrG -- what about now? I 'fessed up, now it's your turn! - Bill Hooker
Not just skim? Title to references? Less than a dozen, certainly. - Björn Brembs
527/277/108 according to Papers (sort by date of import/sort by last opened/count ratings) over the last year. I do read some via Mendeley, ordinary pdf readers or online. Of the papers whose abstracts I read (mostly via RSS), about one fourth ends up in Papers. - Daniel Mietchen
When I was a phD student and postdoc at least 1 a day, now about 1 a week if I am lucky - Kubke
really, really read: 6 a year. - Andrew Lang
Should this convo worry me and/or folks interested in the peer review process? How many papers of those folks *really* read are ones they have reviewed? (Perhaps they aren't counted in these estimates..) What level of reading should be taking place in the peer review process? - Carmen Drahl
None ... that's a separate list - Deepak Singh
Review requires a close reading, but I think everyone has done the same as Deepak and I, that is, considered the two separately. - Bill Hooker
Thanks for the input. I guess I was feeling sensitive this week because I just wrote about a prominent retraction where details in figures and supp info came under fire: - Carmen Drahl
Carmen, that's not to say that peer review actually *does* in practice what it is supposed to in theory; for instance, I suspect that many of the reviews I see as AE at PLoS ONE and BMCRN were written after skimming at best... - Bill Hooker
Sigh. That's a shame. But I'm not sure what to do about that as schedules get busier and journal articles to review get more numerous... - Carmen Drahl
I'm along the Miller/Saunders lines, not counting reviews though - Attila Csordas
600/150/30 = (skim,add to Mendeley / read & discuss > 1hr, add notes to Mendeley / read & re-read carefully, favorite in Mendeley) - Carl Boettiger
Interesting. We seem to be coming up with roughly similar numbers from a fair range of people. I wonder if we all understand the same thing by "close reading" (I suspect we do), and whether there is some kind of cognitive limit on how much of that kind of effort you can put out. - Bill Hooker
I don't know about cognitive limit, but time is certainly limiting. I'm skimming about one per day on average, discussing about 80, and reading and reviewing about 50 or so. - John Hogenesch
Well, I actually was including reviews in my numbers, but it wouldn't skew the numbers much. Bill - I read more web pages and less PDFs these days, maybe a tenth as many PDFs, but those I do read I read more thoroughly, so maybe 100/50 or thereabouts - Mr. Gunn
Y'all might be itnerested in King and Tenopir's work on how much scientists read, and and lots of others. - Joe
Ha, Joe! I was just coming back here to post that! Average researcher reads 250, apparently, mostly found via search & browsing, NOT citation graph (but that could partly be due to poor tools). - Mr. Gunn
Thanks for the link on the jisc article. Don't think I have seen this one before. - Joe
Using RNAi to dissect genetic networks
Iddo Friedberg
CACAO: Community Assessment of Community Annotation with Ontologies -
pretty cool... - Andrew Su
Chris Miller
With the current state of affairs, is it even possible to go straight from grad school to a tenure-track professorial position without doing a postdoc first? Am I wasting my time by applying?
Especially interested to hear from anyone who's been part of a search committee. Is my app just going to get tossed to the bottom of the stack, behind all those with postdoc experience? - Chris Miller
Apparently, if you in fields like CS or Statistics you don't need a post doc to get a TT position. I've always wondered why life sciences insist on post-doc'ing. Is it simply due to a crowded market? - Rajarshi Guha
That's the feeling I get, Rajarshi. The NIH boom created way more applicants than positions. They've gotta go somewhere, so we end up with a glut of postdocs (it doesn't hurt that they're cheap). - Chris Miller
The odds are against you, as you clearly know, but I don't see that it would be a waste of time. What's the worst that can happen: you never hear back, but in the meantime you've practiced your application skills (there's no such thing as too much practice for this), spruced up your CV and thought hard about what you want from a TT position and what you want to do once you get there. I say go for it. - Bill Hooker
Be advised, though, that if you do go straight from grad school to a TT slot, there will be a lot of postdocs who want to punch you in the junk. I mean, a LOT. :-) - Bill Hooker
A different thing to consider is that if you get a TT position, the clock starts ticking. You need to get funding and start publishing pretty quickly and you get a lot more (time-guzzling) responsibilities than a postdoc or graduate student. A postdoc has a lot of negatives, no argument there. But there are positives for the future of an academic career - if you do a (hopefully) short... more... - Mickey Kosloff
You make some good points, Mickey. I'm really just discouraged by the state of affairs and think postdocs are a bum gig for anyone. We somehow think it's okay to pay highly trained experts considerably less than the research techs working under them. The insanity of it all is one of the reasons I'm looking around in industry as well. That all said, if you can't beat 'em... - Chris Miller
You can do industry postdocs -- e.g. I know Genentech offers 'em, and they pay OK. - Bill Hooker
Genentech postdoc positions are highly regarded and I hear it's a great place to do science. However, I think it's still too early to tell if the Roche take-over will eventually change the culture there for the worse. - Mickey Kosloff
Or you can just work in industry ... Which is what I did with zero regrets. - Deepak Singh from iPhone
If you can't beat 'em... forget 'em! ..... :) - Shirley Wu
As someone who had an extremely short postdoc before getting a TT (equivalent) postition, I would strongly recommend postdocing before getting an academic job. The experience is invaluable, the pay is (at least in the UK/AU) not much different, and it is one last chance to devote time to research and not all the administrative crap that surrounds it. If it weren't such a short-term... more... - Anna Croft
Decided difference there between US and Aus/UK -- in the US, as best I can tell, typical postdoc salaries are roughly 50-60% of the starting salary for a TT position. - Bill Hooker
Affordable 'Exomes' Fill Gaps in a Catalog of Rare Diseases
Simon Cockell
I’m giving a lecture next week to the Bioinformatics Masters students here about protein structure prediction. As part of the introduction to this topic, I have a traditional ‘data explosion’ slide, to illustrate the gap between the quantity of protein sequence data available versus the number of solved protein structures in the PDB (hence the [...] - Simon Cockell
Nice work! - joergkurtwegner
Justin H. Johnson
RT @EdgeBio: Views From the Edge blog post about The Growth of DNA Sequencing #ashg2010 #genomics #bioinformatics
Abhishek Tiwari
Fwd: Preserving science: what to do with raw research material? (via
Todd Harris
Informative google doc on making ajax aplications crawlable: #ajax #serversideframeworksaredead
Iddo Friedberg
Justin H. Johnson
Statistical distribution of amino acid sequences: a proof of Darwinian evolution.
Lars Juhl Jensen
The STRING database in 2011: functional interaction networks of proteins, globally integrated and scored -
Lars Juhl Jensen
Deepak Singh
Win-Vector Blog » A Personal Perspective on Machine Learning -
Lars Juhl Jensen
Hidden code in the protein code -
So-called "synonymous" codons aren't, actually. '“One of the bigger implications of our work is that post-translational modifications are actually encoded in the mRNA,” says Kashina. “Coding sequence can define a protein's translation rate, metabolic fate and post-translational regulation.”' - Ruchira S. Datta
Other ways to read this feed:Feed readerFacebook