Wednesday, June 13, 2007

Map your DNA

I recently hiked up the PG&E Trail in Rancho San Antonio Park with my team from work. During the hike we happened to talk about 23andMe. 23andMe is a biotech startup that aims to bring you closer to your genome to provide you "... personal insight into ancestry, genealogy, and inherited traits."

So the question came up: would you have your genome read so that you have a better sense for the health risks you face. That is, would you like to know that you are at a higher risk for prostate/breast cancer, or that you're more likely to get Alzheimer's? Mark was of the opinion that, as long as you understood probability theory and Bayes rule, you should be happy to get a better handle on your risks---the more you know, the better your decisions.

I wasn't so sure I wanted my genome read. Of course, there are situations where it makes sense: if your genome could make predictions with the certainty of a diabetes test or a blood pressure reading and one could treat the resulting condition much the way you can treat diabetes with insulin or hypertension with ACE inhibitors, then I'm all for having my genome read.

But if you get results that say something like "you are 10 times more likely to get cancer or Alzheimer's", then how is that useful? On the one hand, maybe your chances have gone from something like 1 in 10,000 to 1 in 1,000---by no means a certainty. On the other hand, even if you did knew it with certainty, there isn't much you could do other than general healthy living: eat more fruits and vegetables, exercise regularly, keep your brain engaged with crosswords :-), that kind of thing! And on the downside, I can't help but think that knowing you're 10 times more likely to fall prey to a terrible disease only increases your stress levels. I'd rather do without the stress and simply engage in healthy living, blissfully unaware of what fate has in store for me!

Recently the NY Times had an interesting editorial, entitled The Discoverer's DNA, that touched upon this very issue:

Recently, the director of the Human Genome Sequencing Center at the Baylor College of Medicine gave James D. Watson — who with Francis Crick discovered the structure of the DNA molecule — two DVDs that contained the complete sequence of Mr. Watson’s DNA.

And what did Dr. Watson have to say about this:

Dr. Watson has asked not to be told about his version of a gene that has been linked to a predisposition for Alzheimer’s.

The editorial ends with the following:

The possibility of individual genomic sequencing inevitably raises feelings of hope and fear. Both emotions are caused by the same prospect — having a clearer idea of who we are. For now, the most appropriate response is patience.

As DNA science advances and we have a better sense for what companies like 23andMe can do, I might change my mind and have by genome sequenced. But right now, I'm happy to live in ignorance. What would you do?


esmith512 said...

It's been a little over a year since you left your post on the issue and nobody has commented! I have been considering a DNA review as part of a physical and am of the mind that while ignorance is bliss to some, ignorance is unpreparedness to others. We all know we are going to die, but few are hiding from or unreasonably fearful of this certainty. Accordingly, like a road map or other schedule, your DNA allows you to avoid or anticipate predispositions or upcoming events. If I'm predisposed to certain diseases, I'd want to know as soon as possible to afford the greatest lead time to avoid or mitigate the disease factors. I already know I'm predisposed to death, knowing what the likely means of demise are ahead of time allows better planning for a longer life and appropriate planning for that demise. Also, it's simply fascinating to see the source code which created your biological life and facility.

Pandu Nayak said...

I certainly understand that it could be fun to see the "source code" which created ones life---but only if you understood it.

As for knowing whether you are (or are not) predisposed to certain diseases, there is the further question of how certain one is about a change in probability (a Bayesian viewpoint). It would be tragic if there was a prediction that you were much less likely to get a certain disease, only to later find that the prediction was based on partial knowledge and there were other factors that made it more likely that you'd get that disease.

Until DNA-based predictions are much more certain, and the probabilities change in a more significant manner, I'm not sure how useful they are.

/* Google Analytics tracking */