A paper just released in the Lancet describes a thorough and integrated approach to squeezing as much clinically relevant information as possible out of a genome sequence. However, despite a state-of-the-art clinical interpretation pipeline, the major message from the paper is just how far we still have to go before we can make full use of our genetic information.The paper is based on the genome of Stephen Quake (right), which was sequenced using the single-molecule platform developed by Helicos (I wrote about Quake's genome publication at the time). This is a rather curious choice: of all of the genome sequences currently available for analysis, Quake's is one of the least complete and accurate due to the very short reads and high error rates of the Heliscope. It's also interesting to note that at least one of the other authors on the paper - George Church - has a substantially better-quality sequence of his own genome (generated by Complete Genomics) in the public domain.
Nonetheless, Quake's genome it is. The authors throw everything they can at the sequence, bringing in information from databases of both common and rare disease-associated variants and variants affecting drug metabolism, as well as family history and various clinical tests.
There are some genuinely intriguing results: three independent rare mutations in genes associated with sudden cardiac death (although one of these is later shown to be probably benign), and - integrated across the full available set of common risk markers - high lifetime risks for three actionable conditions, myocardial infarction, type 2 diabetes and obesity. Based on Quake's risk predictions, his physicians decided to recommend a lipid-lowering drug (which, incidentally, he would be predicted to respond positively to based on variants in drug-metabolising genes).
That's a tantalising glimpse into the future of personalised medicine using genetic information. Still, it's important to remember what this analysis is missing. Firstly, there's the more than 10% of the genome that can't be accessed at all using Helicos' short-read technology; some of this can't currently be analysed by any of the currently available sequencing technologies (although most would do a substantially better job than Helicos, it must be said).
But more importantly, there are the variants that simply can't be interpreted. This includes virtually everything seen outside protein-coding regions, and the majority of even those variants found inside coding regions. We simply don't understand the biology of most genes well enough yet to be able to predict with confidence whether a novel variant will have a major impact on how that gene operates; and we have an even less complete picture of how genes work together to affect the risk of disease.
That means that the real benefit of whole-genome sequencing over other assays - the uncovering of truly novel or rare genetic variants - has much less of an impact than it should, because in most cases it's impossible to assign function to such variants. Indeed, it's striking in this study that the really compelling, actionable findings - the increased risk of myocardial infarction and metabolic diseases, and the drug metabolism effects - come largely from common variants, most of which would be captured by chip-based assays such as that used by 23andMe. (The two rare variants potentially linked to sudden cardiac death are intriguing, and warrant extra surveillance, but don't yet appear to be compelling risk factors.)
The authors should be commended for their efforts in bringing a wealth of functional annotation and clinical interpretation together for this study, but it's clear we have a lot further to go before we can extract everything of value from a genome sequence.
Medicine & Health
Life Science
Comments
Daniel,
Hmmmm. This will be the quandry for 10 years, 20 max. Time to dig into EnSembl and Zebrafish.....hmmm, conserved region....hmmmm ok, maybe that's bad.....hmmm, what's your cholesterol?
-Steve
Posted by: Steven Murphy MD | April 29, 2010 10:15 PM
Wow, I had no idea we were that far along in this. I figured all the stuff about certain health risks being genetically conditioned was simply based on large-scale studies, not knowledge of the actual specific genes involved. Seems like we'll live to see the day when newborn kids can get some serious medical-prediction stuff (if I'm not mistaken). Neat!
Posted by: Lenoxus | April 29, 2010 11:57 PM
It's an interesting paper. Whole genome sequencing will be interesting for many reasons (particularly I think for human migration history) but it will be the variation scans that will be clinicaly useful whether for SNPs, indels, CNVs etc. Technology is steaming ahead but the biology can't move so quickly. Over 5 years of GWAS we have gone from a few $million per genome to a few $thousand but have not YET made much dent in gene based disease prediction (and I don't believe - re MCK - that 5 years of deep sequencing rare SNP search will be the answer either) - so you're final paragraph is spot on
Posted by: keith grimaldi | April 30, 2010 2:50 AM
I agree! I seriously doubt we will garner much more information about disease risk from genome sequencing than from GWAS studies. Until we have a better understanding of all the epigenetic and other genetic mechanisms (which are likely influenced by environmental exposures) which affect gene expression/translation etc over time, I suspect we will only be making little baby steps in finding specific factors involved in causing high disease risk. Great summary!
Posted by: Susan Colilla | April 30, 2010 9:40 AM
bring it on !
Posted by: 100M1D99M | April 30, 2010 2:59 PM
I'll bet you he has a malaria susceptability gene and if you treated him for lymes disease with doxycyline. He would get fat and have a heart attack from chronic babesiosis. Wilson and Chowning saw this in the Bitter Root Valley in 1905 ---- human piroplasmosis. They got run out of town and the cattle got treated. Good history --real history. Allergy and Immunology? No -- people dying from endemic disease. They saw a tick borne version of malaria back in 05 that still hasn't gotten proper recognition today. Babesia Microti, WA-1, MO-1, CA-1 --- we have very specific tests for those today --- but to convince your gatekeeper to look? You'll get thorazine before doxy. and guess what the nobel prize in medicine was awarded for in 1927? Creating thorazine to treat people with malaria and syphylis.
Thank God for Zyrtec!
Posted by: Rand | May 1, 2010 1:22 AM
I see about 5 patients per week who do not know at least half of their family history. Genetic testing TODAY is cost effective for them; in the future, it will be ILLEGAL not to.
I'm tired of the nay-sayers who want to live in the past: genetic testing is here to stay. Period.
And yes, the marginal utility of a WGS does not merit the massive increase in cost relative to a SNP chip. But that is always subject to revision as the cost curve progresses.
Posted by: Afterthought | May 1, 2010 9:32 PM
Hi Daniel,
Any comment on the MS twin study published in Nature?
Posted by: che | May 5, 2010 3:39 AM
@Steve "....hmmm, what's your cholesterol?"
What Steve advocates is an "occam's razor" of clinical application.
A genetic risk factor is not relevant when the factor itself can be directly tested. For example, if a patient complains of hypertension-like symptoms, you confirm a diagnosis with a sphygmomanometer, and then you treat the hypertension. A genetic test for a risk of hypertension would be redundant as the probability that the patient has hypertension is already "1.0".
However, if the genetic test informed us at some higher resolution about some malfunctioning biochemical pathway which caused the hypertension symptoms, and that information was necessary in some optimal clinical decision ---that would be useful. I'm not aware of any such application in this case.
People hope that DNA will have all the answers to good health. The problem is that those answers are not well translated into the language of medicine ---or for that matter, humans. "Hypertension" in medicine is about as descriptive as "rocky" is in astronomy. ("ICD-9 401.9: Unspecified essential hypertension" is about as descriptive as "a solid, rock-like mass")
Posted by: Andrew Yates | May 10, 2010 10:28 AM
@Afterthought "I see about 5 patients per week who do not know at least half of their family history. Genetic testing TODAY is cost effective for them; in the future, it will be ILLEGAL not to."
*Illegal* not to? This is both terrifying and terrifying. By this same logic, it's cost effective to get your blood pressure. Should it also be illegal not get a regular blood pressure?
Posted by: Andrew Yates | May 10, 2010 10:35 AM
The didn't perform a physical exam, they didn't do a HGBA1c, no PSA, looks like the didn't do a clinical assessment, just played with genome browsers and sequencers. How is that medicine? http://thegenesherpa.blogspot.com/2010/05/personal-genomes-in-clinical-care-quake.html
The answer, it is not.
-Steve
Posted by: Steven Murphy MD | May 10, 2010 10:14 PM
@afterthought
You actually see patients and had a comment like that?
It is truly scary a clinician would suggest it illegal to refuse a test. What's next, Death Panels? Nice.......
-Steve
Posted by: Steven Murphy MD | May 10, 2010 10:16 PM
@ afterthought,
The possibility of using genetic testing to obtain a family history from a patient who can't is something to look forward to someday, but unless I am mistaken, here in the US we can't force anyone to be tested. As it is right now, most people would find the cost prohibitive, and good luck trying to get your HMO to pay for it.
Posted by: Rita Truex | May 13, 2010 7:03 PM