The DNA Network

Maybe Bird Brains Aren't So Bad [adaptivecomplexity's column]

Posted: 28 Aug 2008 04:27 PM CDT

Along with chimps, orangutans, elephants, and dolphins, some birds are smart enough to recognize themselves in the mirror. We're naturally inclined to think of mammals as the creatures most likely to have a developed sense of self, but it turns out that magpies are also quite self-aware.

How do you know when an animal recognizes itself in the mirror? You need some way to tell whether an animal thinks a mirror image is another animal, or whether the animal recognizes itself.

Researchers get at this problem by marking the animal in some way and watching how it behaves. In this case, a group of German researchers stuck a yellow sticker on the bird, in a place where the bird would only notice the sticker when looking in a mirror:

Image from Prior, et. al, PLoS Biology 6(8): e202

The test birds, seeing the sticker in the mirror image, frequently started searching their own bodies for the mark - a sign that these birds knew they were looking at an image of themselves.

These results add to the evidence that, like some mammals, certain bird species have brains that can carry out more complex cognitive functions. Most animals don't have much use for the kind of self-awareness exhibited in this mirror test. But self-awareness can be useful, the German researchers argue: in the social environment of these birds, more self-aware individuals might be better at using past experience to predict the behavior of other birds in their social group.

Posted: 28 Aug 2008 11:35 AM CDT

Mumps was a common childhood disease when I was a child. We grew up learning that it was better to get mumps as a child because getting it as an adult would make you sterile. No doubt that idea arose from symptoms like swollen glands, swollen testicles, etc. When I looked in PubMed though, I couldn't find much data on sterility (at least not easily).

I did find data on hearing loss.

Read the rest of this post... | Read the comments on this post...

Why look at genetic association one locus at a time when you can do several loci at a time? [Yann Klimentidis' Weblog]

Posted: 28 Aug 2008 09:55 AM CDT

This looks pretty cool. They propose a method of doing association studies by considering several SNPs at once instead of one SNP at a time. Even though this makes biological sense, it seems like the analysis would get messy and cumbersome. They seem to address these issues and analyze their method on previous data sets. They use a "Bayesian inspired" method called: stochastic search maximisation algorithm.
I'll need to read this paper more closely to see exactly how this method compares to previous single-SNP analyses, one of which, they describe in the paper, is on T2D.

S imultaneous Analysis of All SNPs in Genome-Wide and Re-Sequencing Association Studies
Clive J. Hoggart, John C. Whittaker, Maria De Iorio, David J. Balding
PLoS Genetics 4(7)

Abstract: Testing one SNP at a time does not fully realise the potential of genome-wide association studies to identify multiple causal variants, which is a plausible scenario for many complex diseases. We show that simultaneous analysis of the entire set of SNPs from a genome-wide study to identify the subset that best predicts disease outcome is now feasible, thanks to developments in stochastic search methods. We used a Bayesian-inspired penalised maximum likelihood approach in which every SNP can be considered for additive, dominant, and recessive contributions to disease risk. Posterior mode estimates were obtained for regression coefficients that were each assigned a prior with a sharp mode at zero. A non-zero coefficient estimate was interpreted as corresponding to a significant SNP. We investigated two prior distributions and show that the normal-exponential-gamma prior leads to improved SNP selection in comparison with single-SNP tests. We also derived an explicit approximation for type-I error that avoids the need to use permutation procedures. As well as genome-wide analyses, our method is well-suited to fine mapping with very dense SNP sets obtained from re-sequencing and/or imputation. It can accommodate quantitative as well as case-control phenotypes, covariate adjustment, and can be extended to search for interactions. Here, we demonstrate the power and empirical type-I error of our approach using simulated case-control data sets of up to 500 K SNPs, a real genome-wide data set of 300 K SNPs, and a sequence-based dataset, each of which can be analysed in a few hours on a desktop workstation.

Bill Clinton tells America to look at the facts [The Daily Transcript]

Posted: 28 Aug 2008 08:56 AM CDT

Instead of vague platitudes, Clinton uses empiricism to make his points in last night's speech - America needs to look at what happened over the past 8 years. From last night's speech:

Read the rest of this post... | Read the comments on this post...

New blood [genomeboy.com]

Posted: 28 Aug 2008 08:20 AM CDT

Okay so maybe it’s not so new since the book’s been out for several months…Anyway, my review of Masha Gessen’s Blood Matters: From Inherited Illness to Designer Babies, How the World and I Found Ourselves in the Future of the Gene, appears in the current issue of Nature Genetics:

Throughout this remarkable hybrid of a book—part memoir, part science journalism, part narrative nonfiction—Gessen demonstrates both her independence and her willingness to tweak dogma, whether it comes from guilt-ridden postmodernists or didactic medical professionals. Her sometimes conflicting goals are to discover what is possible for herself as a breast cancer ‘previvor’ and to quench her reporter’s compulsion to document and understand the genetic landscape as it shifts beneath her feet.

Root canal treatment [Mailund on the Internet]

Posted: 28 Aug 2008 05:44 AM CDT

I’ve just come back from the dentist. I had a root canal treatment for my toothache.

That was lots of fun.

I think I’ll go and get some nice paper cuts now; maybe pour some lemon in them….

GINA is Wrong? [Think Gene]

Posted: 28 Aug 2008 04:15 AM CDT

Did anyone actually read GINA (Genetic Information Nondiscrimination Act) before lauding it for protecting the public or condemning it as over-regulation?

From GINA at The Library of Congress:

Prohibits a group health plan from requesting or requiring an individual or family member of an individual from undergoing a genetic test. Provides that such prohibition does not: (1) limit the authority of a health care professional to request an individual to undergo a genetic test; or (2) preclude a group health plan from obtaining or using the results of a genetic test in making a determination regarding payment. Requires the plan to request only the minimum amount of information necessary to accomplish the intended purpose.

So, according to this:

Health plans can require the results of a genetic test to make a decision regarding payment.
Assuming having one’s genome available to make medical decisions makes health care more effective and efficient, premiums cannot be adjusted to account for the savings.

Um, isn’t this is the worst for both effective health care and patient fairness? Doesn’t this mean that health plans can simply demand any genetic test when deciding payment (and hold that decision hostage to get that test… and what if a decision is urgent)? Doesn’t this mean that health care will be institutionally less effective because rather than instituting a single, preventive genomic test to be consulted to make better medical decisions for one’s entire life(which GINA makes illegal), instead, patients will only get ad-hoc and myopic genomic tests as demanded by health plans for payment decisions? Won’t this make health care less effective while creating a new market for inefficient, interventionist specialty genomic tests?

So, what happens when a project like the Coriell Personal Medicine Collaborative shows that using genomic personalized health care makes health care better and cheaper?

I’m embarrassed that I relied on the hype surrounding GINA and didn’t read the law to form my own opinions.

Further, if I’m right about this (I want to consult a lawyer), I’d be deeply embarrassed for the journalists and industry representatives who should have noted this rather than publishing fluffy, feel-good fed PR. Hypothetically. I’m going to get a legal opinion before I start stomping around, naming names, and posting links!

I guess Neanderthals weren’t stupid after all… [Mailund on the Internet]

Posted: 28 Aug 2008 02:56 AM CDT

Actually, I have no particular reason to believe that Neanderthals should be dumber than Homo sapience. Well, they were around for thousands of years with very little if any development in the kind of tools they used (or left behind, in any case) while we have managed to improve on tool design quite frequently.

At least lately, I don’t know how fast we were innovating in the stone age.

In any case, what I wanted to go with this is to this page that describes a study of Neanderthal and Homo sapience tools and conclude that they were equally efficient.

Does this mean that Neanderthals were as smart as us? I guess it doesn’t necessarily, but at least they were as technically advanced.

How did we outcompete them, exactly, and why are they gone?

The ‘Ubiquitous’ web [business|bytes|genes|molecules]

Posted: 28 Aug 2008 12:13 AM CDT

Image via Wikipedia All of you know about it already, but I shall happily add to the noise. Last evening I had one of those “Holy S**t” moments. Was sitting in a coffee shop, catching up with the days news, when I saw a flurry of activity on Friendfeed around Ubiquity. Turns out Ubiquity is a new project by Mozilla Labs, which for want of a better description is like Quicksilver for the browser, a mini command line available with an Alt-space.

Ubiquity for Firefox from Aza Raskin on Vimeo.

Ubiquity is still young, and may never catch on although I have a feeling it will, at least among the geek crowd, and being a Firefox plugin = low friction. But you can see the promise right away. You can, very quickly, using simple commands access search, Wikipedia, maps, insert material into documents, send email, etc.

Here are some examples

But that’s just a start. It doesn’t take a leap of faith to see entire vocabularies being created to support certain data types and activities. An early example comes from Rajarshi Guha, who very quickly rustled up a couple of commands. Maybe we can have a repository somewhere someday for a set of standardized commands in bioinformatics/cheminformatics

The example above isn’t quite working perfectly, but you can see what we can do, and this is just the tip of the iceberg. There are a lot of mashups possible, including getting related papers, targets, compounds, structures, and being able to package them up and email them, or put them in a document, and as our web gets a little more structured, I can imagine myself sitting inside that command line and following an entire graph of information that streams through, ready to be manipulated and used for something even more interesting.

The web is a moving target, our browsers are moving targets, our capabilities to manipulate are moving targets, and efforts like Ubiquity show us a glimpse into the future. Will be fun being part of that future (and blogging about it)

Update: I should have known Pawel would take a stab. An even cooler example (Sorry Rajarshi, I like proteins better)

Markets being analyzed.... [The Gene Sherpa: Personalized Medicine and You]

Posted: 27 Aug 2008 08:41 PM CDT

I was recently invited to participate in a survey for concierge practices. Since we run Genomic Medicine/Personalized Medicine practices in New York and Connecticut I agreed. Then the questions...

[[ This is a content summary only. Visit my website for full links, other content, and more! ]]