Finding the Genetic Links in Common Disease: Caveats of Whole Genome Sequencing Studies
Writer and Reporter: Stephen J. Williams, Ph.D.
In the November 23, 2012 issue of Science, Jocelyn Kaiser reports (Genetic Influences On Disease Remain Hidden in News and Analysis)[1] on the difficulties that many genomic studies are encountering correlating genetic variants to high risk of type 2 diabetes and heart disease. At the recent American Society of Human Genetics annual 2012 meeting, results of several DNA sequencing studies reported difficulties in finding genetic variants and links to high risk type 2 diabetes and heart disease. These studies were a part of an international effort to determine the multiple genetic events contributing to complex, common diseases like diabetes. Unlike Mendelian inherited diseases (like ataxia telangiectasia) which are characterized by defects mainly in one gene, finding genetic links to more complex diseases may pose a problem as outlined in the article:
- Variants may be so rare that massive number of patient’s genome would need to be analyzed
- For most diseases, individual SNPs (single nucleotide polymorphisms) raise risk modestly
- Hard to find isolated families (hemophilia) or isolated populations (Ashkenazi Jew)
- Disease-influencing genes have not been weeded out by natural selection after human population explosion (~5000 years ago) resulted in numerous gene variants
- What percentage variants account for disease heritability (studies have shown this is as low as 26% for diabetes with the remaining risk determined by environment)
Although many genome-wide-associations studies have found SNPs that have causality to increasing risk diseases such as cancer, diabetes, and heart disease, most individual SNPs for common diseases raise risk by about only 20-40% and would be useless for predicting an individual’s chance they will develop disease and be a candidate for a personalized therapy approach. Therefore, for common diseases, investigators are relying on direct exome sequencing and whole-genome sequencing to detect these medium-rare risk variants, rather than relying on genome-wide association studies (which are usually fine for detecting the higher frequency variants associated with common diseases).
Three of the many projects (one for heart risk and two for diabetes risk) are highlighted in the article:
1. National Heart, Lung and Blood Institute Exome Sequencing Project (ESP)[2]: heart, lung, blood
- Sequenced 6,700 exomes of European or African descent
- Majority of variants linked to disease too rare (as low as one variant)
- Groups of variants in the same gene confirmed link between APOC3 and higher risk for early-onset heart attack
- No other significant gene variants linked with heart disease
2. T2D-GENES Consortium: diabetes
Sequenced 5,300 exomes of type 2 diabetes patients and controls from five ancestry groups
SNP in PAX4 gene associated with disease in East Asians
No low-frequency variant with large effect though
3. GoT2D: diabetes
- After sequencing 2700 patient’s exomes and whole genome no new rare variants above 1.5% frequency with a strong effect on diabetes risk
A nice article by Dr. Sowmiya Moorthie entitled Involvement of rare variants in common disease can be found at the PGH Foundation site http://www.phgfoundation.org/news/5164/ further discusses this conundrum, and is summarized below:
“Although GWAs have identified many SNPs associated with common disease, they have as yet had little success in identifying the causative genetic variants. Those that have been identified have only a weak effect on disease risk, and therefore only explain a small proportion of the heritable, genetic component of susceptibility to that disease. This has led to the common disease-common variant hypothesis, which predicts that common disease-causing genetic variants exist in all human populations, but each individual variant will necessarily only have a small effect on disease susceptibility (i.e. a low associated relative risk).
An alternative hypothesis is the common disease, many rare variants hypothesis, which postulates that disease is caused by multiple strong-effect variants, each of which is only found in a few individuals. Dickson et al. in a paper in PLoS Biology postulate that these rare variants can be indirectly associated with common variants; they call these synthetic associations and demonstrate how further investigation could help explain findings from GWA studies [Dickson et al. (2010) PLoS Biol. 8(1):e1000294][3]. In simulation experiments, 30% of synthetic associations were caused by the presence of rare causative variants and furthermore, the strength of the association with common variants also increased if the number of rare causative variants increased. “
Figure from Dr. Moorthie’s article showing the problem of “finding one in many”.
(please click to enlarge)
Indeed, other examples of such issues concerning gene variant association studies occur with other common diseases such as neurologic diseases and obesity, where it has been difficult to clearly and definitively associate any variant with prediction of risk.
For example, Nuytemans et. al.[4] used exome sequencing to find variants in the vascular protein sorting 3J (VPS35) and eukaryotic transcription initiation factor 4 gamma1 (EIF4G1) genes, tow genes causally linked to Parkinson’s Disease (PD). Although they identified novel VPS35 variants none of these variants could be correlated to higher risk of PD. One EIF4G1 variant seemed to be a strong Parkinson’s Disease risk factor however there was “no evidence for an overall contribution of genetic variability in VPS35 or EIF4G1 to PD development”.
These negative results may have relevance as companies such as 23andme (www.23andme.com) claim to be able to test for Parkinson’s predisposition. To see a description of the LLRK2 mutational analysis which they use to determine risk for the disease please see the following link: https://www.23andme.com/health/Parkinsons-Disease/. This company and other like it have been subjects of posts on this site (Personalized Medicine: Clinical Aspiration of Microarrays)
However there seems to be more luck with strategies focused on analyzing intronic sequence rather than exome sequence. Jocelyn Kaiser’s Science article notes this in a brief interview with Harry Dietz of Johns Hopkins University where he suspects that “much of the missing heritability lies in gene-gene interactions”. Oliver Harismendy and Kelly Frazer and colleagues’ recent publication in Genome Biology http://genomebiology.com/content/11/11/R118 support this notion[5]. The authors used targeted resequencing of two endocannabinoid metabolic enzyme genes (fatty-acid-amide hydrolase (FAAH) and monoglyceride lipase (MGLL) in 147 normal weight and 142 extremely obese patients.
These patients were enrolled in the CRESCENDO trial and patients analyzed were of European descent. However, instead of just exome sequencing, the group resequenced exome AND intronic sequence, especially focusing on promoter regions. They identified 1,448 single nucleotide variants but using a statistical filter (called RareCover which is referred to as a collapsing method) they found 4 variants in the promoters and intronic areas of the FAAH and MGLL genes which correlated to body mass index. It should be noted that anandamide, a substrate for FAAH, is elevated in obese patients. The authors did note some issues though mentioning that “some other loci, more weakly or inconsistently associated in the original GWASs, were not replicated in our samples, which is not too surprising given the sample size of our cohort is inadequate to replicate modest associations”.
PLEASE WATCH VIDEO on the National Heart, Lung and Blood Institute Exome Sequencing Project
https://www.youtube.com/watch?v=-Qr5ahk1HEI
REFERENCES
http://www.phgfoundation.org/news/5164/ PHG Foundation
1. Kaiser J: Human genetics. Genetic influences on disease remain hidden. Science 2012, 338(6110):1016-1017.
2. Tennessen JA, Bigham AW, O’Connor TD, Fu W, Kenny EE, Gravel S, McGee S, Do R, Liu X, Jun G et al: Evolution and functional impact of rare coding variation from deep sequencing of human exomes. Science 2012, 337(6090):64-69.
3. Dickson SP, Wang K, Krantz I, Hakonarson H, Goldstein DB: Rare variants create synthetic genome-wide associations. PLoS biology 2010, 8(1):e1000294.
4. Nuytemans K, Bademci G, Inchausti V, Dressen A, Kinnamon DD, Mehta A, Wang L, Zuchner S, Beecham GW, Martin ER et al: Whole exome sequencing of rare variants in EIF4G1 and VPS35 in Parkinson disease. Neurology 2013, 80(11):982-989.
5. Harismendy O, Bansal V, Bhatia G, Nakano M, Scott M, Wang X, Dib C, Turlotte E, Sipe JC, Murray SS et al: Population sequencing of two endocannabinoid metabolic genes identifies rare and common regulatory variants associated with extreme obesity and metabolite level. Genome biology 2010, 11(11):R118.
Other posts on this site related to Genomics include:
Cancer Biology and Genomics for Disease Diagnosis
Genomics-based cure for diabetes on-the-way
Personalized Medicine: Clinical Aspiration of Microarrays
Late Onset of Alzheimer’s Disease and One-carbon Metabolism
Genetics of Disease: More Complex is How to Creating New Drugs
Mitochondrial Metabolism and Cardiac Function
Pancreatic Cancer: Genetics, Genomics and Immunotherapy
Quantum Biology And Computational Medicine
Personalized Cardiovascular Genetic Medicine at Partners HealthCare and Harvard Medical School
Consumer Market for Personal DNA Sequencing: Part 4
Personalized Medicine: An Institute Profile – Coriell Institute for Medical Research: Part 3
Whole-Genome Sequencing Data will be Stored in Coriell’s Spin off For-Profit Entity
I actually consider this amazing blog , âSAME SCIENTIFIC IMPACT: Scientific Publishing –
Open Journals vs. Subscription-based « Pharmaceutical Intelligenceâ, very compelling plus the blog post ended up being a good read.
Many thanks,Annette
I actually consider this amazing blog , âSAME SCIENTIFIC IMPACT: Scientific Publishing –
Open Journals vs. Subscription-based « Pharmaceutical Intelligenceâ, very compelling plus the blog post ended up being a good read.
Many thanks,Annette
I actually consider this amazing blog , âSAME SCIENTIFIC IMPACT: Scientific Publishing –
Open Journals vs. Subscription-based « Pharmaceutical Intelligenceâ, very compelling plus the blog post ended up being a good read.
Many thanks,Annette
I actually consider this amazing blog , âSAME SCIENTIFIC IMPACT: Scientific Publishing –
Open Journals vs. Subscription-based « Pharmaceutical Intelligenceâ, very compelling plus the blog post ended up being a good read.
Many thanks,Annette
I actually consider this amazing blog , âSAME SCIENTIFIC IMPACT: Scientific Publishing –
Open Journals vs. Subscription-based « Pharmaceutical Intelligenceâ, very compelling plus the blog post ended up being a good read.
Many thanks,Annette