History in our Genes

Via the Wired Science Blog, I found this story about a study of variations in the human genome among different populations. While this has been done before, the new study shows how much data we can get just by looking at people's genes.

"The novel finding is the depth of the resolution we've gone to," said National Institutes of Health neurogeneticist Andrew Singleton, co-author of one of the papers in Nature. "This really lets you start moving towards locating individuals geographically. Previously, we've been able to look at the genome and say, 'This part is from Africa, this is from Asia. Now we can look past that and say, 'It's from this part of Africa or Eurasia.'"

Continued Singleton, "We can use these data to look at other areas of the genome that might have been under particular pressure for survival, and go from there to figuring out what the pressure is. One area that was highlighted was the genes responsible for digesting lactose. In countries where there's milk consumption, that one particular haplotype that allows more efficient lactose digestion has arisen."

They've not only been able to identify populations based on the genome alone, but they've also managed to model how humanity spread around the globe. Our history is encoded in our genes. How cool is that?

And all this was done using only the genomes of about a thousand people. Imagine what will be possible once we have even more data. And on the biology side, we will be able to repeat the same analysis for animals or plants. These are truly exciting times.

The full article that appeared in Science can be read here.

Evolving Evolution

Okay, that title was too easy a pun. I apologise. But Wired blog had a couple of posts about where evolution (as in, "the theory of...") may be headed. The name for this extension of evolution is complexity theory. I bet that's going to be confused with chaos theory a lot.

What is complexity theory? Let's let Wired explain:

Not a religiously inspired revision -- intelligent designers need not apply. Nobody suggests that genetic mutation and natural selection aren't responsible for the evolution of birds from reptiles or humans from tree-swingers.

But a growing number of scientists do say that neo-Darwinian evolution doesn't explain certain jumps in biological complexity: from single-celled to multicellular organisms, from single organisms to entire communities.

The jumps -- saltations, in complexity parlance -- appear to be non-linear emergent phenomena, the result of networked interactions that produce self-organization at ever higher levels. From this perspective, Darwinian evolution is a mechanism of a higher universal law, perhaps even a variant on the second law of thermodynamics.

There's something that will strike fear into the hearts of mathematicians and computer scientists everyhwere: non-linear emergent phenomena. Everybody knows that that is the last thing you want to be modelling.

Personal fears aside, this seems like a reasonable extension. It is important to find out how certain networked properties emerge, whether they be multi-cellular organisms, or human societies.

Now, some people, including me before I read this article, might think that classic evolution might be all we need to explain these jumps. Just because we haven't figured it out yet, doesn't necessarily mean that we need a new framework, right?

Maybe, but the second Wired article that caught my eye gives more evidence that our current view of evolution may limit our ability to explain certain phenomena.

When Guy Hoelzer runs computer simulations of organisms living in the modeling equivalent of a featureless plain, he sees them break into different species -- even though there's no reason for natural selection to take place.

That preliminary but tantalizing finding hints at some larger phenomenon driving the mechanisms of neo-Darwinian evolution. Hoelzer thinks the phenomenon is self-organization: combine energy with complex networked interaction and order will emerge.

As with all experiments based on simulation, you have to take this with a grain of salt, but it's certainly more fodder for complexity theory.

One thing I don't quite agree with is that they try to relate it back to the second law of thermodynamics. ("The entropy of an isolated system not in equilibrium will tend to increase over time, approaching a maximum value at equilibrium.") That seems doubtful to me (speciation is entropy?) and also a bit premature. Save the unification with physics for when you've worked out the details of your theory.

It's Started Already

Remember the "telepathic" DNA story? Well, what'd I tell you:

DNA Found to Have "Impossible" Telepathic Properties

[...]No one knows how individual DNA strands could possibly be communicating in this way, yet somehow they do. The “telepathic” effect is a source of wonder and amazement for scientists.

So we've moved from a "telepathy-like quality", that can be explained, to "impossible telepathic properties". I predict that we'll reach alien influences in two more steps.

I love Bad Science

Let me clarify: I hate bad science, but I love Bad Science. Ben Goldacre cuts right through the bull and exposes charlatans, cranks and clueless journalists. In his latest column, he dissects a collection of recent articles that are full of bad science.

I know I’m wrong to care. On the BBC news site “crews were hopeful the 20m cubic litres of water could be held back and not breach the dam wall”. And that’ll be a struggle, since “cubic litres” are a nine-dimensional measuring system, so the hyperdimensional water could breach the dam in almost any one of the five other dimensions you haven’t noticed yet.

Seriously, go read it.

Interesting Article, Bad Title

LiveScience had an article about a recent discovery in genetics made by some researchers in Maryland. They found out that DNA sequences that are the same are more likely to cluster together than those that are different.

Curiously, DNA with identical sequences of bases were roughly twice as likely to gather together as DNA molecules with different sequences.

[...]

The electrically charged chains of sugars and phosphates of double helixes of DNA cause the molecules to repel each other. However, identical DNA double helixes have matching curves, meaning they repel each other the least, Leikin explained.

It's an interesting discovery, although it remains to be seen how useful it's going to be. But that's not the reason I'm mentioning it here. What made me laugh was the title they gave the article:
DNA Molecules Display Telepathy-like Quality.

How long until the first New Age healer picks up on this and claims he can heal your DNA by telepathy?

Woe is PhD

So I'm looking for a PhD place. Ideally, I'd like it to be at my current university, as the other universities in the UK that are good in my field are mostly located in London, and I don't like that city much. Nor can I really afford to live there.

I'd like it to involve Bioinformatics, but not require any wetlab work that I would have to do myself. There should also be scope for applying Machine Learning techniques. There's two areas of research that interest me. One is work in genetics, such as gene regulation modelling, or protein structure and function prediction. The other is to model biological systems at the macro level and predict how changes in the environment influence animal population size or plant growth.

I could also see myself doing a straight Machine Learning project without any Bioinformatics involved, but with slightly less enthusiasm.

I'd like to have a supportive supervisor, who I can talk to before I start my PhD and who will advise me on how to write up my project proposal. He or she doesn't need to be an academic superstar, but a fair number of puplications and at least some amount of recognition in either Bioinformatics or Machine Learning would not go amiss.

I'd like to get the chance to teach during my PhD, either tutorials or even lectures.

A scholarship would be helpful. If I don't get one, my parents could help, but I'd like to be able to support myself for once.

And tomorrow, I will meet with a potential supervisor who might be able to offer the place that has most of these characteristics. (I'm not sure about the teaching yet.) Fingers crossed!

Google's Newest Invasion

Having cornered the search market, taken over Youtube and struck fear into the hearts of publishers, Google is licking its lips and looking for new targets. Next stop, Wikipedia.

The potential Wikipedia killer app that Google is developing is called knol, and it has of course sparked much discussion in the blogosphere, with bloggers outbidding each other in who can come up with the wittiest pun. (My favourite: Google Sets its Guns on the Grassy Knol). But puns aside, should Wikipedia be worried? Should we?

Yes and no. You have to remember that Google does not always suceed. Remember Google Video? That didn't take off. Google News? I don't know any News junky who uses that. And I don't see people rushing to pick up Google Talk.

Knol operates on a completely different model from Wikipedia. Instead of having a page for a topic that everybody can edit if they think they know better than the original author, in Knol, the author controls the content, and other users can only suggest changes. Also, you could have more than one "knol" on each topic. Google assures us that more popular (and hence, presumably, more accurate) knols will float to the top, but will they really?

Of course, you have a similar problem in Wikipedia, but it's mitigated because you can correct information: In knols, it seems you can only decide if you like the information or not.

Then there is the problem of orphaned topics: What happens if a "good" knol is abandoned by the author? Can we reuse it to start a new knol? Or is that information frozen in time forever, neither to be reused nor updated?

There are tons of problems that could bring knols to its knees. Of course, there are also reasons why it could succeed: There's less risk of vandalism like Wikipedia has seen, and there are more incentives for people to contribute (Google has agreed to share ad revenue if knol owners let them place ads on their pages).

Whether or not knol succeeds, I believe that the two models are different enough that they could even exist side-by-side. After all Encarta and the Encyclopedia Britannica both sold copies, didn't they?