Was the data for the ClinSeq SNP project obtained from patients or general population

[CNVboy]

SNP database from 1000G or NHLBI projects are based on general population, or presumably "healthy" people.

A long-time confusing question for me is, ClinSeq SNP project is based on just patients?

I look into their website, and I'm sure that the volunteers involved in the project DEFINITELY include "patients" or at least involving various types of syndromes, and in their genome research paper they also reported some rare missense mutations which they think associated with say, some abnormal cardiovascular phenotypes.

And also you can find, in dbSNP, there'll always be some rare mutations absent from 1000G/NHLBI, but present in ClinSeq.

Then my questions are: Does ClinSeq involve ONLY patients, or also involve general population? And for those rare mutations ONLY found by ClinSeq, can we view them as highly-likely deleterious or pathogenic?

Thanks!

[jxchong]

The NHLBI ESP subjects include what you might consider "unhealthy adults" (have ordinary adult-onset diseases like heart attacks, etc) as well as "healthy" controls.

ClinSeq also has similarly healthy/unhealthy adults.

From the paper:
ClinSeq participants are being selected to represent the spectrum of atherosclerotic heart disease using the Framingham score (Wilson et al. 1998) to balance accrual. The accrual target is three groups of 250 participants each who have a Framingham 10-yr coronary artery disease (CAD) risk of <5%, 5%–10%, and >10%, respectively, and an additional group of 250 participants who have a diagnosis of CAD based on a history of a myocardial infarction, coronary artery bypass graft surgery, stent placement, or other re- vascularization procedure.


Presumably a coronary artery disease risk of <5% is pretty good (healthy) and coronary artery disease doesn't generally imply what I would consider highly deleterious/pathogenic/Mendelian.

I would definitely NOT assume that a variant found in ClinSeq but not 1KG/ESP to be highly deleterious!!! Much more likely just a rare variant or resulting from difference sequencing methods, etc. IIRC ClinSeq did Sanger sequencing of a number of genes, so you might expect Sanger to pick up certain variants that NGS would not.