Please, do you mean polygenic deseases or true chromosomal defects?

It would be difficult to give a simple answer. Multiple defects or defects in multiple genes can be associated with a particular disease or cancer. By analyzing whole genome data, we may be able to correlate those defects with a particular disease. I think there are a lot of such whole genome data accumulated in recent years.

I have seen quite a few studies looking at gene-gene interactions, which I think is the main thing you are asking about (not just additive effects of multiple variants, but statistical interactions between different variants). My research group has been interested in looking at gene-gene and gene-environment interactions. You could look up some of the papers by my colleagues on interaction between DAT, DRD4, and CHRNA4 genetic variants and prenatal exposure to maternal smoking (main authors include RD Todd and RJ Neuman). They found that a certain combination of variants in these genes plus prenatal maternal smoking greatly increased the chance for a severe form of ADHD. I think it can be tricky to study interactions among multiple genes at once due to power issues, but if you have a plausible hypothesis regarding why there might be a gene-gene interaction, it can be useful to investigate the possibility using that small set of genes. In the case of DAT, DRD4, and CHRNA4 genes, they all have potential to interact to influence dopamine neurotransmisison, and the exposure to nicotine could also affect this since binding of nicotine to neural nicotinic acetylcholine receptors can affect the amount of dopamine released.

See the following reference for further description of this particular GxGxE model:

Todd RD, Neuman RJ.

Am J Med Genet B Neuropsychiatr Genet. 2007 Dec 5;144B(8):971-5.

Christian, thank you so much for your question. I'm looking to see if people are developing tools that analyze large datasets to mine for genetic interactions that may be unknown.

It was interesting that after reading your question, I looked up the term "polygenic diseases" and ended up finding this definition from MedicineNet: "Polygenic disease: A genetic disorder that is caused by the combined action of more than one gene. Examples of polygenic conditions include hypertension, coronary heart disease, and diabetes. Because such disorders depend on the simultaneous presence of several genes, they are not inherited as simply as are single-gene diseases."

I guess I was very surprised that what they listed were very common disorders that also have so many environmental components. What is the burden of proof of this sort of association? Christian, were these the conditions you were thinking about? Do you know some papers where they were talking about relative risk in anything being called a polygenic disorder? Does the same combination of gene defects always produce the disorder?

I think Angela's answer was more what I was looking for as far as having people looking at genes that would have clear biological associations where a snp that generates a dysfunction in one gene product may influence another gene product that might produce an unusual problem that may manifest in disease.

I've spent some time looking at results in the Strings database looking at possible interactions in some undiagnosed people, but wondering if there is a new science opening up at this place, and who the players are. Thank you, Angela, I will look for those papers.

Hua Xiao , it is my impression that a lot of our datasets generated by GWAS studies are more for heavily funded conditions, such as the ones listed in that MedicineNet definition. I am more committed to trying to find answers in places where doctors know the patient is sick, and may not even have a diagnosis, but the physician may have no tools for figuring out what issues the patient has that may stem from an unfortunate turn of genetics.

Some of my interest is coming from my myoadenylate deaminase deficiency project because I have about 65 people now on my facebook group for that disorder and their issues really are complex and not so much about muscle. Is their elevated adenosine interacting with regulation of other genes? What accounts for the diversity in how that disorder manifests in this group where we are looking at people selected by their genotype instead of their phenotype?

Certainly, this ability to start with the genotype is a big shift. What is the best way to study the contribution of any other genetic components?

Thanks so much, everybody, for your thoughts, They are greatly appreciated.