Colbert on DNA

Great episode of the Colbert Report tonight - "DNA: Could It Happen To You?".

Mostly focused on familiar bits of genetics - the Innocence Project, National Geo's Genographic Project, which has collected DNA samples from more than 100,000 folks for geneological purposes. Introduced, though, by a nicely done spoof 1950s film on where DNA is from (from Jesus!).

This follows Colbert's recent interview with Craig Venter, aka DNA to Z, where a lot of the audience didn't really seem to get the import of what Venter was saying.

Thing is, Colbert is very good at spotting the zeitgeist and parodying it before others spot it in the first place - so the fact that he's twice made fun of DNA means that genetics is very much of the moment.

Welcome to the Microbiome

Much swell stuff out of this weekend's SciFoo conference. But probably the coolest thing that I saw was Jonathan Eisen's presentation on the Microbiome. As the genome is to genes, the microbiome is to microbes - the comprehensive catalog of all the microbes in an ecosystem - in Eisen's talk, the human body. As Eisen describes it on his blog:

In essence the human microbiome is the sum collection of all the microbes found in or on people. The human microbiome has become an important research field because the microbes that live in and among us play critical roles in human disease and health. An important aspect of this is the idea that microbes can be and are beneficial. For example, in the gut the normal microbes help with digestion and nutrient absorption as well as protect from infection. In addition, a variety of diseases (e.g., IBD, Chrohns) seem likely to be caused by disruption in the normal microbial flora. In general, it seems likely that other ailments, like autoimmune diseases, allergies, etc will be found to have a connection to disruptions in the beneficial microbes that live among us.

The cool thing here is it makes perfect sense and is at once both comprehesible and profound to most people. Everybody knows we've got stuff in our gut helping us digest stuff (the booming probiotics trend testifies to this). But the idea that these organisms are essential, not simply beneficial, and that they may be instrumental in our understanding of disease, immunology, and so forth - well, that's simply stunning.

As Eisen explains, the NIH just created the Human Microbiome Project - an effort to genetically sequence and catalog all of these microbes and start to suss out what they do (the analogy to the Human Genome Project and the ensuing efforts to understand the function of human genes is deliberate and inescapable). The NIH has designated the project part of its NIH Roadmap for Medical Research, a fast track for developing new technologies.

The trick, of course, is actually finding out what's in and on us. There are, Eisen estimates, 100 times as many microbic cells affixed to a body as there are in that body - meaning there are likely thousands of microbes in a given person, and there is a tremendous variety from person to person.

It was a great talk especially because of the quality of the questions - and of the questioners. At one point Freeman Dyson was offering a quip and Drew Endy was urging Eisen on and George Church was sitting in the back, comparing notes with various folks.

The Non-Genetics of Cancer

A clarifying poll from a couple UK organizations I've never heard of - Cancerbackup and Genes Reunited - shows that the overwhelming preponderance of people (in the UK, at least) think that cancer is largely hereditiary/genetic. In fact, this story says, 90 percent of cancers are entirely random. The biggest risk factor for cancer is not family history, as 60 percent of those polled believe, but age.

SciFoo

At SciFoo. If the rules allow, I'll be posting when I can.

The Ultimate Medical Device

  My latest story in Wired is all about the race to develop the ultimate diagnostic tool, a handheld gizmo that will might diagnose infectious disease based on the presence of just a blip of DNA material. It so happens, though, that last week I was in Seattle and visited Microsoft Research to hear about what they're up to in health care. One conversation turned me on to what may be the real ultimate tool in medicine - and there are already 3 billion of them in use around the world. It's the cell phone.

Here's the story: A couple years ago, in late 2005, Microsoft put out a small RFP for health care programs on "Digital Inclusion Through Mobile and Wireless Technologies" - the idea being to sponsor programs that used cheap mobile and wireless technologies to solve a broader problem. The awards were very small - just $1.2 million spread over 17 winners - but the idea wasn't to fully fund these programs so much as to lend Microsoft's credibility so that local NGOs and other donors might be brought on board. (Microsoft is also contributing some devices and software to the programs).

The 17 winners were chosen last year, and they're just now getting ready to implement their programs. The programs span a wide range of geography and topics - from mesh networks for security in Beunes Aires to using PDAs and smartphones to enable microfinance in Uganda.

But to me, the most intriguing programs are those that use simple, low-cost cellphones to create potent, robust health care networks. In Botswana, for example, where HIV infections are rampant, they are creating a Healthcare Information Service that taps exists cellular networks to disseminate HIV information. The idea is that education may be the greatest medicine, but it's difficult to get out into local communities.

In Pakistan, researchers from Carnegie Mellon University are helping develop a similar health-information network, this one using speech-recognition technology to deliver info on a broad range of health issues. Speech recognition is necessary because much of the medical services in Pakistan aren't delivered by MDs but by "lady health workers," a battalion of 100,000 government trained first-line workers who deliver primary medicine in communities. Many of these LHWs, though, are illiterate, making speech recognitition a neccessity. These are just simple cellphones, mind you, not fancy devices. And in many of the 17 programs, they're all that's needed to draw together communities and exchange information that can fundementally improve health conditions.

Especially cool is how Microsoft is running the program - already they've brought all the project heads together to share their results and compare notes, allowing some great cross-fertilization of problems and solutions. "If the project in Uraguay is using this ambient network, then it's relevant to the project in Viet Nam," says Tom Healy, program manager for Microsoft Research's external research and program efforts.

Makes you wonder what kind of cool stuff we could be doing here in the US with all the devices we already have surrounding us...

The Ultimate Diagnostic Device

The new issue of Wired has a story I wrote on a new breed of molecular diagnostic tools - small, portable devices that detect telltale segments of DNA, that could be used at the point-of-care (from the doctor's office to a Saharan village) and be a major weapon against infectious disease, from XDR TB (which is why the WHO is interested) to anthrax (which is why the DoD is interested). I focus on Akonni Biosystems, a small startup out of Maryland that's crafting one promising microarray tool based on technology developed at Argonne National Lab. Here's the gist:

There is a new crop of diagnostic tools on the horizon, portable devices that can detect infectious disease with a degree of accuracy that measures up to that of lab-based cultures. The approach blends the values of the technology sector, in which products live or die based on how well they scale toward cheaper, simpler versions, with the priorities of global public health, which holds that if a solution isn't cheap and simple to use, it may as well not exist. The result is an emphasis on cost, speed, size, and simplicity. It's a formula that could change the way infectious disease is detected and treated.

The Magic Bullet for Hospital-Borne Infections? Q-Tips.

A counter-intuitive story in Friday's NYTimes about hospitals that are successfully fighting drug-resistant infections. At the Pittsburgh VA Hospital, cases of drug-resistant infection have dropped 78 percent since 2001, when they instituted certain disease prevention measures.

Those measures, tellingly, are not dousing the hospital in more antibiotic solutions, but rather the simple act of swabbing incoming patients nostrils and other basic steps. Patients are quarantined if they have drug-resistant bacteria - typically methicillin-resistant Staphylococcus aureus, or MRSA, which I've written about recently (I love how the Times calls the bacteria "virulent" - an interesting adjective for bacteria, considering its root). Other Pittsburgh hospitals have focused on the simple act of hand-washing, which has been known to be a deterrent for infection since Semmelweis pinned it to maternal mortality in 1847 (the Freakonomics guys had a nice column about this a few months back).

The Times story doesn't get into this, but there's a reason why Pittsburgh may be ahead of the game. In 2005, the Pennsylvania Hospital Cost Containment Council, an independent state agency, issued a report that found hospitals reported 19,154 cases of infections that originated in Pennsylvania hospitals, a rate of 12.2 per 1,000 cases at a cost of $3.5 billion. Pennsylvania was the first state to collect and report information about hospital-acquired infections.

In other words, it's easier to act when you have data upon which to act.

The Onslaught of Genomic Associations

Missed this last week: a nice report from NPR's Science Friday on genome-wide association studies with some bigshots including Lawrence Brody from the National Human Genome Research Institute. Picking up on the recent reported associations for breast cancer, restless leg syndrome, and so on, they try to explain what to make of these associations. The second half of the program gets into the first genome-assocation study of infectious disease. Researchers from the Center for HIV/AIDS Vaccine Immunology found that certain genetic variants can affect vulnerability to HIV infection (some people who are infected with HIV are asymptomatic, in that their immune system appears to control the virus). In other words, depending on your genes, you can be more vulnerable to certain infectious diseases.

Daniel Koshland Jr

Just heard that Donald Koshland, a biochemist at UC Berkeley, passed away earlier this week at age 87. The former editor of Science magazine, Koshland was an expert in enzymes and an early supporter of genetic engineering. I had the opportunity to hear one of Dr. Koshland's last lectures at Berkeley, and he struck me as a great teacher and innovative thinker (indeed, that lecture inspired a forthcoming feature in Wired magazine). Addressing a room full of usually inattentive undergraduates, Koshland was engaging, insightful, and enthusiastic; at lecture's end, he had a crowd of students around the podium asking him more questions (including me). Here's a fine obituary from the San Francisco Chronicle.

Re: The Fat-as-Contagion Study

Much in the news about this study in this week's New England Journal of Medicine about obesity being contagious. Two tiny comments:

1) Nice to see that the researchers used retrospective data from the Framingham Heart Study, my favorite study. Framingham is one of the longest ongoing epidemiological studies ever; as the authors note, the study started in 1948, when 5209 citizens of Framingham, Massachusetts, were signed up. The study moved into a second generation of 5124 subjects in 1971, and a third generation of 4095 people in 2002, all direct descendents of the original cohort.

To date, Framingham has led to more than 1,200 published research findings (that number is through 2004, so is probably much higher by now). And that doesn't include citations, which must number in the tens of thousand. Much of what we now know about cholesterol, high blood pressure, the risks of cigarettes, and heart disease all stems out of the good citizens of Framingham. The study is now moving into genetics, with over 5,000 DNA samples. I've always thought that somebody could write a great book about Framingham.

2) I particularly appreciate how the NEJM got physicist Alberto Barabasi to write an editorial on the social network theory underlying the study (you may've heard of his book Linked). He makes a very good point about how network effects are increasingly relevant (and increasingly better understood) for health:

The growing interest in interconnectedness has brought into focus an often ignored issue: networks pervade all aspects of human health. One example of this trend involves social networks and their impact on the spread of obesity or pathogens — from influenza to the severe acute respiratory syndrome or the human immunodeficiency virus.

Noting that these effects will have ever greater relevance in a genetic context, Barabasi goes so far as to suggest that we're at the dawn of a new field; he dubs it "network medicine."