Announcing BioBits

Lately, I’ve been struggling with the fact that I’ve been writing for two different audiences on my weblog. The first group consists of my family, friends, and people that have similar tastes in politics and culture. The second group are fellow scientists and techies who may be interested in what I’ve been up to in lab, or what new bioinformatics tools I’ve been using.

In the interest of pleasing both audiences and allowing you to more easily sort the wheat from the chaff, I’m moving my science writing to chrisamiller.com/science while maintaining my personal blog right here. I want to emphasize that this isn’t about separating my personal and professional lives, as that’s a near impossibility in this age of the internet. Rather, it’s about offering focused content to just the people that want to hear it.

I’m also making a resolution to post on a near-daily basis over there, to get that site off on the right foot. Some days it may just be a code snippet, or a good quotation, but I’ll try to keep the content flowing with some more meaty posts about bioinformatics, article reviews, and thoughts about scientific culture.

Bwaaahaha!

A fed-up student asks the physics prof who’s going over kinetics in protracted detail, what good is all this? What will I ever use this for? To which the professor, not even looking up from where he’s writing on the board, says “This stuff saves lives.”

The student balks for a second and then gets belligerent, demanding to know how first-year physics saves lives. The professor doesn’t even turn around, saying “it keeps idiots like you out of med school.”

via

The Ripple Effect

On AskMe, someone posted about their struggle to find meaning in life after deciding that atheism was for them:

I realize that, without “something behind everything”, it doesn’t matter one iota (speaking from a selfish perspective here) whether I build great things or just sit on my couch and rot, whether I live to be 100 or die tomorrow. It will matter to some, but not to many, and not for long.

I decided to respond:

Surely, you’ve known someone who has changed your life immeasurably for the better. This may be your parents, the guy who pulled you from that burning car, or the teacher who encouranged you to make use of your talents. I bet you wouldn’t say that their contribution meant nothing.

I promise that I’m not getting all hippie on you here, but both the love and hate you express tends to propagate through society, starting with the people you interact with every day. No, really – there has been behavioral research showing that this phenomenon exists.

Furthermore, your life choices, like quitting smoking, or losing weight make it more more likely that your peers will do the same.

As James Fowler puts it (as quoted in the above link):

“Everyone always tells me that this research is so depressing and that it means we don’t have free will. But I think they’re forgetting to look at the flipside. Because of social networks, your actions aren’t just having an impact on what you do, or on what your friends do, but on thousands of other people too. So if I go home and I make an effort to be in a good mood, I’m not just making my wife happy, or my children happy. I’m also making the friends of my children happy. My choices have a ripple effect.”

Now go out and start some ripples.

Double-slit experiment

Your daily dose of quantum weirdness.

QOTD

Clay tablets were heavy and tended to shatter.
Papyrus crumbled but at least it was flatter.
Parchment was handsome but prone to decay.
Paper with acid will just melt away.
Magnetic disks are erased in a trice.
Something that lasts, really lasts, would be nice.

— David Drake

QOTD

If any student comes to me and says he wants to be useful to mankind and go into research to alleviate human suffering, I advise him to go into charity instead. Research wants real egotists who seek their own pleasure and satisfaction, but find it in solving the puzzles of nature.

— Albert Szent-Györgi (1893-1986) U. S. biochemist.

Lessons for article recommendation services

Today someone proposed the creation of a sub-reddit where scientists could recommend papers to each other. While it’s a nice thought, I can almost guarantee that it’s going to be a failed effort. There are already sites like Faculty of 1000, which try to use panels of experts to recommend good papers. In my experience, they mostly fail at listing things that I want to read.

The main reason such sites are useless is that we scientists are uber-specialized, so what you think is the greatest paper ever will likely have very litle interest for me. It’s not that I don’t want to read about cool discoveries in other fields, it’s just that I don’t have time to. Until they invent the matrix-esque brain-jack for rapid learning, I have to prioritize my time, and my field and my work will always come first.

There are only two systems I’ve found that work well. The first are recommendation systems based on what you’ve read in the past, and what your colleagues are reading. CiteULike, for example, recommends users that have bookmarked similar papers to you, and perusing through their libraries gives me an excellent source of material. The other quality source of recommendations is FriendFeed, where I can subscribe to the feeds of other bioinformaticians with similar interests, and we can swap links to papers and comments about those papers.

Both of these systems are all about building micro-communities, with a focus that you can’t achieve in larger communities like Reddit. In this way, it’s sort of like a decentralized version of departmental journal clubs, or specialized scientific conferences. Any site that ignores the value of creating this type of community is pretty much doomed to failure from the start.

Again with Big Genomics

Since my last post, the question of why we need big genome centers has been bouncing around in the back of my mind. One commenter on that post said:

The reason is quality. A large center can (and in my experience does) perform much more quality analysis and produces more sequence more cheaply.

While I don’t disagree, with sequencing price dropping dramatically, couldn’t this be done just as well by one of the dozen or so commercial operations that now offer sequencing? They’re all doing business on a large scale and this have just as much ability to offer quality product. Why should the NIH continue to subsidize something when it’s being offered at reasonable rates from the corporate sector?

So no, I don’t believe that big genome centers will be any better than industry at preserving read quality. As I said before, they will still be the ones helping to get next-gen sequencing platforms off the ground, but that’s research, not providing a service. So what other advantages might they offer?

The idea that I’ve been kicking around is that they might offer a critical mass of talent in the same place. Instead of meeting twice a year at conferences, scientists work right down the hall from each other, and can thus bounce ideas off each other more quickly and easily. This probably also helps prevent duplication of effort. If you know Professor X next door is working on problem Y, you won’t be wasting your time trying to beat him to the punch. Instead, you’ll be contributing and collaborating with him.

That said, I think physical proximity and openness about research are quickly becoming less relevant, with the advent of online science. We can develop close-knit communities of researchers online and do more of our work in the open, which will actually help prevent scooping.

So how about it – what am I missing? What are your ideas for why big genomics centers can or will remain viable?

The fade of big genomics

In a recent post, David Dooling asks why genome centers are forced to release their data early, when other smaller labs with a sequencing machine aren’t. In responding to some of the comments on his post, David had this to say:

Most of what large genome sequencing centers are doing now are not the type of projects that were done five to ten years ago, i.e., not the type of projects that fall under Bermuda/Fort Lauderdale. The goals and end results of many of the projects are no different than those of single investigators.

I agree completely, and think it begs the question: What do we need the sequencing centers for, then?

For many years, the centers were essentially acting as contractors. The NIH said: We’ll give you $50M, you give us a bunch of genomes. Back then, big centralized projects were necessary, because producing reasonable amounts of sequence required a lot of infrastructure and the economies of scale. Those times are quickly coming to an end.

It’s no secret that sequencing is becoming a commodity. In just a year or two, any lab will be able to get hundreds or thousands of genomes sequenced, either in-house, or by sending DNA off to a company. The few remaining genome centers have tried to keep themselves relevant by focusing more on the science, but it’s getting to the point where most of that science could be done anywhere.

Let me be very clear that I’m not anti-genome center (in fact, I’m affiliated with one here at BCM). I think that they still have a tremendous role to play in developing new sequencing techniques and bringing them into the mainstream. They won’t fade away completely, but I think that we will see a marked reduction in the size and scope of the projects that they tackle.

Really, I think of this process as analogous to the early days of computing. Supercomputers at universities were the only game in town for a long time. They played their part and did amazing work, but the cost-reduction and decentralization of the computing infrastructure was ultimately a good thing. We’re seeing the same thing happen in biology, and we’ll all be better off because of it.

Air bubbles

Have you ever left a glass of water sitting out overnight, then found little air bubbles all over the sides the next morning? Yeah, so did this guy, and he wanted to know why. Since I spent ten minutes answering him, I figured I’d repost part of the answer here:

To put it simply, solubility is the amount of solute that a solvent can hold in solution (got all that?). It might be easier for you if you think about stirring sugar into water. You add a little, it stirs right in and all the sugar dissolves. If you keep adding more and more, the water eventually won’t be able to hold any more, and the sugar will start precipitating out to the bottom of the glass.

Water also dissolves other things, like gases. In addition to trace amounts of minerals, some of the nitrogen and oxygen from our air is dissolved into the water.

Now here’s the important part: The solubility of water isn’t constant, and is affected by temperature and pressure. When these things change, the amount of stuff that your glass of water can hold changes.

Hot water will be able to dissolve more sugar than cold water. This fact is often demonstrated by making rock candy in grade school science lab. Heat water up to boiling, and stir as much sugar in as you can. Then, stick a popsicle stick in the solution and let the water slowly cool. Over a day or so, crystals will start to form.

What happened? Well, a few sugar molecules stick to rough spots on the stick. These are called sites of nucleation. A few more molecules of sugar will start to stick to those, then a few more, and soon visible crystals will start to form. All this happens because the solubility of the water has decreased as it cools. So the water can’t really hold all the sugar anymore, and the sugar desperately wants to get out of solution.

The same principle is at work with your glass and air bubbles. There are dissolved gases in the water, and imperfections in the glass serve as nucleation sites. That’s why you see bubbles sticking to the edge of the glass. Over time, if the bubbles keep growing, the force compelling them to rise will overcome the forces sticking them to the side of the glass, and the bubbles rise to the top and disperse. To speed this up, you can tap the glass to help the bubbles break free from their nucleation site, and a bunch of them will rise at once.

Bad graphs and bad reporting

Several blogs, including GOOD and the New York Times Economix blog have been reporting on obesity, and using this graph as evidence:

The authors of the graph (and the blogs) claim it shows that countries whose populations eat more slowly have less obesity. I took one look at that graph and said “Bullshit”.

The points are scattered all over, and if there is a correlation between the two factors, it’s a very weak one. To prove this, I re-created the graph in Open Office by approximating their data points. I then did a logarithmic fit of the data, which produces a graph that is very close to the original. So far, so good:

When you fit lines to data, there are ways to measure how well that line describes the data, by using something called the Coefficient of determination (R-squared). This value ranges from 1, which means the line fits perfectly, all the way down to zero, which means the line is basically arbitrarily drawn.

If you look at my graph, you’ll see that the R^2 value of this line is 0.18. That means the fit is very weak and should pretty much be ignored. Even though the line draws your eye and makes you believe the relationship, there is very little correlation here.

Now, this doesn’t mean their conclusion is wrong. There very well may be a link between speed of eating and obesity. However making that inference from this data would (and should) get you laughed out of any serious medical or scientific meeting. This is a case of bad graphing, bad reporting, and plenty of blame to go around.

QOTD

“For a long time, science had not been in the forefront. It was faith or science, take your pick. Now we’re saying that science is the answer to our prayers.”

–Rep. Nancy Pelosi

Scientifically Scooped

I’m working on a manuscript now, and I find myself feeling kind of jumpy. Everytime I skim the journal RSS feeds and see a keyword related to my project, I get a little jolt until I finish reading the title/abstract and confirm that no, in fact, they did not beat me to the punch.

I realize logically that there are a huge number of problems in science, and that there’s very little reason to think that someone is going to scoop me. Still, it’s terrifying to think that my entire last 6 months of work could be rendered obsolete by one paper that duplicates my findings first, or even worse yet, does it more elegantly!

That said, I’m still glad to be working on this article and hopefully, getting a little closer to graduation.

Refrigerators as a risk factor?

The latest issue of PLOS One contains an article titled Crohn’s Disease and Early Exposure to Domestic Refrigeration. The major result?

This study supports the opinion that CD is associated with exposure to domestic refrigeration, among other household factors, during childhood.

Say whaaaa? Refrigeration is causing disease? Well, okay, that’s not quite what the researchers are saying. (Remember correlation does not equal causation). So what are they saying? Well, to understand, we first need a quick primer on Crohn’s Disease:

We all have trillions of bacteria living in our gut, and for the most part, it’s a peaceful co-existence. They eat a little of the food, and help us digest some of it, and we provide a warm safe environment for growth. When a person gets CD, though, their immune system suddenly begins seeing these little buggers as a threat and starts attacking them. The result is a chronically inflammed intestinal tract, and some nasty GI problems. No one knows exactly why some humans get it, but we do know that it’s a mix of genetic and environmental factors.

This study aimed to find out what environmental factors might be associated with the disease. After comparing people with and without CD, refrigeration turns out to be one of several domestic technologies associated with increased prevalence of the disease. So, it could be that refrigeration itself is the cause, or it could be a standard-of-living thing.

CD is in an interesting class, because it’s a mostly first-world disease – that is, incidence is highest in people with higher incomes and higher standards of living. Most diseases that we think of (say, AIDS or malaria) are associated with the poor. It’s easy to see why that’s so – contaminated water, poor access to doctors, limited education, etc.

So why the first-world association? Well, one hypothesis is that it’s all about hygiene. Specifically, it says that our germophobic lifestyles are to blame. Since we don’t get exposed to as many microbes on a regular basis, when we see an unfamiliar one, our immune system is more likely to go all kung-fu on it.

Based on the results of this and one other study, the authors float a second idea, which they call the “cold chain” hypothesis. The gist is that while most bacteria grow very very slowly at cold temperatures, there are a few species that can grow better than others. When we buy cold foods at the supermarket, they’re likely to be on ice for a few days in the supply chain, then a few more days in our fridges, giving these cold-resistant bacteria time to grow up.

Of course, this all hinges on the idea that these cold-tolerant bacteria are uniquely responsible for CD. I’ve read a fair amount of the literature on CD, and frankly, I’m not convinced. Several individual types of bacteria have been proposed as causative in the past, and so far, none of those ideas have panned out. (MAP, enteroadherent E. coli). It seems more likely that many different types of bacteria can take advantage of a susceptible patient.

I’m also skeptical because, as they state in the paper, ownership of televisions, cars and washing machines were also associated with the disease, nearly to the same extent. Given all these confounding factors, I’m more inclined to believe that the association with refrigerators is just a byproduct of increased standard of living. In that case, this study can be seen as just another data point cementing CD’s status as a first-world disease and perhaps even lending support to the hygiene hypothesis.

QOTD

“The species of whale known as the black right whale has four kilos
of brains and 1,000 kilos of testicles. If it thinks at all, we know what it is thinking about.”

–Jon Lien

Blogging Darwin: Chapter One

As promised, I’m following along with Blogging the Origin and reading Darwin’s On the Origins of Species. I’m a few days behind John, and but here are my thoughts on Chapter 1: Variation Under Domestication

• The chapter is mainly a huge selection of examples that prove a basic premise: There are variations between members of a species, and humans, by selecting for desirable traits, can alter the species over time. Sure, it’s obvious now (and was sort of obvious then), but the rest of Darwin’s case relies strongly on this premise.
• It’s clear what Darwin’s strategy is, right off the bat. Club the reader over the head with example after example, to make sure that the concepts are clear and that the assertions are substantiated. Pigeons, Pears, Dogs, Sheep – he covers them all. You can feel that he knows in controversial territory and he wants his story to be bulletproof. He often states the anticipated arguments against his work and pre-emptively addresses them.
• Though it’s a little long-winded at times, I wish science had more of these detailed and comprehensive explanations and less 3-page papers in high-profile journals that omit nearly all the background and methodology.
• It’s fascinating to read Darwin’s speculations on the mechanisms of inheritance. Remember, back in 1859, modern genetics was still completely unknown, and there were plenty of half-baked ideas floating around. (I’m talking about you, Lamarck). It’s amazing to see how much he got right (inheritance, hints about recessive traits), and amusing to see what he got wrong. Considering that Mendel’s work wasn’t discovered for a few decades, and that DNA wasn’t shown to be the mechanism of inheritance until the next century, it’s pretty darn good.

Next up: Chapter 2: Variation Under Nature

Blogging Darwin

John Whitfield has a confession to make: He’s a PhD in evolutionary biology, and has never read “The Origin of Species”.

I can relate: I bought a copy years ago, intending to read it, but just never made the time. As I’m doing my PhD work in genomics and deal with evolutionary ideas all the time, I also feel a little sheepish. John intends to rectify this situation, so he’ll be reading the book a little at a time and blogging about each section. Fantastic idea! Hopefully there will be some good discussion in the comments as he progresses.

I’m planning on dusting off my copy and reading along. Find yourself a copy and do the same, but do it quick – he’ll be posting thoughts on the introduction soon.

The Chaos Inside a Cancer Cell

Our lab’s latest paper was featured in the New York Times today!

A few quick notes:

• All props go to Oliver Hampton, who did the lion’s share of the work to get this paper out. (I’m the third author on the paper).
• I realize that to scientists, the NYT isn’t as exciting as getting published in a big-name journal, but it sure seems to mean a lot more to my non-scientist friends and family!
• The featured picture was generated using the Circos software package. (open source, GPLed)
• The paper is open-access and available as an advance publication here.

Finally, a president who respects science

It’s time we once again put science at the top of our agenda and worked to restore America’s place as the world leader in science and technology.

- President-elect Barack Obama in his weekly address

It’s about time we heard that from Washington DC, and it’s great to see that he’s backing it up by appointing several well known and respected scientists to key positions. From my perspective, it’s especially exciting to have Eric Lander on the President’s Council of Advisors on Science and Technology. He’s one of the biggest names in genomics and should do a great job of informing this White House about the power and potential of personalized medicine.

Update: How did I miss Harold Varmus? He’s a Nobel-laureate cancer researcher. As someone working on cancer genomics, this lineup makes me very happy.

The BBC gets it wrong

Wow, this BBC article about breast cancer and embryo screening gets it wrong big-time.

[H]e said that, in this case, not carrying the BRCA1 gene would not guarantee any daughter born to the couple would be unaffected by breast cancer because there are other genetic and environmental causes.

Bzzzzzzzt. Sorry. This science reporter fails the test, and does so repeatedly throughout the article.

Everyone has the BRCA1 gene, and it actually helps prevent cancer! See, BRCA1 is involved in the process of DNA damage repair, which reduces the number of mutations in the genome. A properly functioning BRCA1 protein helps stop cancer before it starts.

What confused the reporter was that only some people have BRCA1 genes with mutations. When one or a few base pairs in the BRCA1 gene are altered, it can change the resulting protein and cause it to malfunction. These mutations are what confer a higher risk of breast cancer.

At least one scientist interviewed in the article gets it right:

Dr Alan Thornhill, scientific director of the London Bridge Fertility, Gynaecology and Genetics Centre, said: “While the technology and approach used in this case is fairly routine, it is the first time in the UK that a family has successfully eliminated a mutant breast cancer gene for their child.

It probably could have been more clearly said, but he does state that it’s the mutant copy of the gene that’s being eliminated, not the gene itself.

You may say that I’m playing semantics here, but this is crucially important stuff, as the article makes very clear. People are going to have to begin making serious decisions about risk factors affecting themselves and their children. It’s hard enough to convey this information when we get the facts right. Muddying the waters with bad reporting only makes the job tougher.

Update: Props to the BBC for changing the article. I don’t know if it was my contact form submission or someone else’s, but they’ve updated the article to make it more correct. That said, shouldn’t they acknowledge the change somewhere on the page? A footnoted erratum would go a long way towards improving transparency.