Comments on: Tilting At Genomic Windmills

By: Carman

Carman — Fri, 17 Apr 2009 16:00:09 +0000

Robert – If you don’t think schizophrenia is wide spread in the animal kingdom, then you clearly have not met my cat.

By: Orson

Orson — Fri, 17 Apr 2009 14:01:22 +0000

Chalk up another win for the conservative medical correspondent from the London Telegraph, Dr. James LeFanu, author of ”The Rise and Fall of Modern Medicine.” He argued this point years ago.

By: Jonathan

Jonathan — Fri, 17 Apr 2009 13:32:44 +0000

That Goldstein perspective was actually painful to read. The science was fine, but did NEJM fire their editors or something?

By: Robert Grumbine

Robert Grumbine — Fri, 17 Apr 2009 13:21:50 +0000

Sorry for the lack of detail, but iirc, Nature carried a science article on a related point. Namely, that the association studies are finding far less effect from the detected genes than there is scope in the gene-related disease. That is, after deciding that some disease was 20% genetic, and running the association study to blame a gene, the blamed gene could only account for 2% of the genetic contribution. (I’m making up the numbers here, but the general drift is right — the identified genes were small players even in diseases that were known to be genetic.)

The 1,000 genes involved in schizophrenia … wow! We have only about, what, 23,000 genes? And the vast majority have to be involved in basic cellular stuff we share across the entire eukaryotic empire — how to metabolize, how to make multiple cells stick together, how to build tissue types, etc, etc. Wouldn’t that many genes being involved in schizophrenia point to it being an illness that we should be able to find across a fair swath of at least the animal kingdom?

By: Tuatara

Tuatara — Thu, 16 Apr 2009 16:52:43 +0000

Yep, evolution and epigenetics is a really dangerous mixture of topics that a lot of folks go on crazy speculative rants about. I get totally frustrated trying to think about epigenetics in the context of population genetics. That being said, there may be a lot of deleterious (albiet ephmeral) epigenetic polymorphism floating around human populations that likely cause diseases from time to time.

By: Ashutosh

Ashutosh — Thu, 16 Apr 2009 15:34:50 +0000

In the context of epigenetics you may want to read Larry Moran's generous words about it and George Johnson. Also, it's interesting that ancient Indian medicine was very much aware of the extreme variety of the same disease occurring in different individuals. That's why my grandfather who was a well known doctor- both modern and Ayurvedic- used to say things like "I prescribed different asthma medications for these two people because his asthma is not the same asthma as the other one's. "Personalized medicine" studied in an empirical manner was known to ancient Indian medicine.

By: Tuatara

Tuatara — Thu, 16 Apr 2009 14:03:04 +0000

At first it does seem surprising that deCODE would make such an argument. However, with all the new technology to look at genome-wide patterns of chromatin modification and methylation, epigenetic phenomena is likely to become a very important component of personalized medical genomics.

By: TomJoe

TomJoe — Thu, 16 Apr 2009 13:50:31 +0000

Wait a second … is deCODE actually arguing for epigenetics? If so, that sort of undermines their whole business model. deCODE is in the business of doing sequencing, and if they’re now arguing that non-DNA related changes are resulting in disease they won’t be able to find it no matter how hard they sequence. Or am I missing something?