Newsweek Reports on "Genetics False Promises"
Perhaps this report on the utility of DNA/genetics/genomic research will help move autism research away from the genetics-heavy approach of the last decades and into more usable, actionable research for biologically based treatments. It's time to re-evaluate the current funding model within the autism grant community.
Read the full article in Newsweek.
By Sharon Begley
DNA As Crystal Ball: Buyer Beware When it comes to predicting risk of disease, Alzheimer's genes—and others—strike out.
...Yet, as the JAMA study shows, there are serious doubts about how useful genomic information is going to be, outside of a few rare applications such as the ability of a child with leukemia to metabolize chemotherapy, one of the earliest attempts to pair genomics with medicine...
Personalized medicine has many high-profile partisans, such as Francis Collins*, director of the National Institutes of Health, who made the case for the field in his recent book. (emphasis and asterisk ours.) Nevertheless, second thoughts are clearly setting in as a result of studies like those I outlined above. Last year, geneticist Steve Jones of University College London wrote in The Daily Telegraphthat despite the billions of dollars that governments, industry, and foundations have poured into genomics and personalized medicine, "the mountain has labored and brought forth a mouse," one that will have little effect on how medicine is practiced, let alone predicting someone's risk of disease.
* You may recall a post written about Dr. Collins by David Kirby titled. NIH Director Francis Collins Blames Resignation of Top Health Official from Autism Panel on “Tension and Lack of Trust
I strongly suspect that the problem has to do with the way the genes are found in the first place.
For the sake of argument, say you pick a single gene, carry out an experiment with a huge group of subjects and discover a large statistical correlation with some specified disease, and your statistical analysis shows a 99.9% probability that this result was not due to chance. Then we would all be inclined to accept this correlation as indicating a real causation (to be precise, we could say it is real with a probability of 99.9%). However keep in mind that this means there is a 1 in 1000 probablility that the result was due to chance and therefore meaningless.
Now you carry out this experiment, testing not only one gene, but simultaneously testing 5000 genes. Then you can expect to find 5 false positive results, i.e., 5 genes for which you can now falsely claim that there is a large correlation with the disease. Then you go on to develop drugs to inhibit the gene and and then you sell the drug and basically try to make as much money as you can before anyone attempts to replicate your result, because when they do they will discover that in fact it doesn't hold up. The more subjects you had in your original study, the harder it will be to arrange such attempts at replication.
Ever notice how, after finding these genes, the researchers go on to state that they have no idea how the gene actually works to produce the disease? And of course, having found the gene in such a manner, it is quite unlikely that the gene has any effect on the disease at all.
But this is exactly the point. One ought to start with the actual molecular effects of a particular gene, theorize based on that about what diseases that gene might influence, and then carry out a study looking for a correlation between that particular gene and those particular diseases. If under THOSE circumstances you find a large correlation, you can have a lot of confidence in such a result. But I think this is not how things are done in the majority of cases these days...
Actually computing the expected number of false positives for these studies makes a nice (and for a mathematician, easy) math problem which someone ought to carry out and publish.
Posted by: Eric | May 23, 2010 at 04:00 PM
Ah yes. I remember all those miracle cures that were to come from the mapping of the human genome, now of course we have the stem cell yahoos ready to canabalize every embryo from every species on the planet in the name of science. Most of the people that scream the loudest for these miracle cures know the least yet surprisingly have the greatest say. This will not be the first or last time you see this type of hysteria by the psuedoscientist who strangely almost always come from the same political ideology.
Posted by: WILLIE | May 22, 2010 at 06:41 PM
Maybe Newton's Law applies here. To every action there is an equal and opposite reaction.
The stronger the research into genetics the greater the likelihood the true answer lies in too many pharmaceuticals.
Posted by: Cynthia Cournoyer | May 20, 2010 at 10:50 PM
Maybe this will lead to more people understanding that very little has come from genetic study. What has come from the discovery of the genetic link for Down syndrome?
When ever I hear talking heads announce they have found some kind of gene that will lead to solutions for Autism, I believe they are talking about preventing the birth of any fetus that carries a genetic marker for any imperfection that may occur. They will go to hell.
Posted by: K Fuller Yuba City | May 20, 2010 at 07:45 PM
Yep, it's time for things to swing the other way again. I'd like to see a study re: BPA levels in pregnant women/breastfeeding women and newborns as markers for intestinal permeability to be used as a predictor of adverse reactions to vaccines. Now that would be meaningful. Research out of Toulouse, France late last year, showed that BPA levels only 1/10 as low as those publicly declared safe, caused problems in the intestines. It was shown through in utero tests and through breastfeeding exposure. i.e. the environment causes intestinal problems which then causes new susceptibilities to additional environmental toxins (vaccines?). Hey, is this how measels virus can end up in the stomach? Shoot it into a baby who already has gut permeability at birth and it leaks from the backside of the stomach into the digestive system itself, creating a circular cycle of ongoing destruction until the child's body cannot take anymore and regression starts. Or maybe some children are so so susceptible that those are the kids that have ASD signs right from birth. Hey Dr. Wakefield, do you have your next project yet?
Posted by: Jenny | May 20, 2010 at 11:56 AM
I wish the title would have been "Genetics False Prophets."
Posted by: paul5of6 | May 20, 2010 at 10:10 AM
When I was going to school, the big question was which played the bigger part in behavior, and disease; genetics, or the environment.
Looks like it going to be the environment!
Who knew!
Posted by: Benedetta | May 20, 2010 at 08:09 AM