This builds on prior question http://www.biostars.org/post/show/11990/genotype-data-and-exome-sequencing-data-sensitivity/
If I care only about variants covered by dbSNP with an rsXYZ number and assuming I could do two things on a single clinical patient:
A) genotyping with some array (e.g., Affymetrix with 1M SNPs)
vs.
B) exome sequencing, and filtering the resulting VCF file to only variants covered by dbSNP and of high quality
Which one will be more accurate?
I know that B can give me more data about indels and such but my clinical-implications-of-genome-data-knowledge-base gives me clinical advice only if it is a previously known variant. So I am more after a method which is most accurate. In fact I only need to know if the sequencing will be at least as accurate as genotyping. (assuming "average" sequencing vendor and pipeline)
2 answers
If you are interested in more than just exome variants then you will have to do whole genome sequencing. If you want it to be as accurate as a genotyping array you will need quite a lot of sequencing depth to get sufficient coverage across all the same regions as covered by an Affy or Illumina 1M assay. Eventually this will be feasible and cost-effective with sequencing. But, I don't think we are there yet. Since you are only interested in established dbSNP variants, it seems the genotyping array makes a lot more sense for you and will save you a ton of money. Of course, this assumes that the current genotyping arrays have all the variants you are interested in.
Definitely. The sequence data would be better for "other purposes". If you can afford to do WGS sequencing to 30-40X then it will probably be about as accurate and comprehensive as the arrays and give you a whole lot more. But, be prepared to spend a couple orders of magnitude more and don't forget to factor in the extra cost/effort you will need for analysis.
These two papers are relevant for the question:
http://www.ncbi.nlm.nih.gov/pubmed/19420381
http://www.plosgenetics.org/article/info:doi/10.1371/journal.pgen.1000832
Obtained from this discussion http://seqanswers.com/forums/archive/index.php/t-7641.html
which seem to indicates that WGS is better then genotyping.
Log in to answer this question.
-
Searching for human genotyping data thats not public
written by Bioinformatics-man •
-
Basic difference between haplotype variant callers
written by sofie_carolina •
-
Job: We'Re Hiring! Postdoctoral Fellow - Genomic Informatics, Mount Sinai Hospital, Toronto, Canada
written by mpoushin •
-
What Agilent Interval Files (.Bed) Should I Use For Exome Variant Calling With Gatk?
written by newDNASeqer
-
What The The Most Commonly Used Databases For Clinical Annotation Of Genomic Results?
written by user56
-
Tutorial: Genotype And Snp Calling Review Papers
written by Istvan Albert
-
How To Filter Vcf Files From 1000 Genomes Release V3.2010-11 (Alternative Source)?
written by user56
-
If I Order Wgs From A Vendor (Eg. Illumina), What Files Do I Get Back From Them?
written by user56
-
Sequencing Result: Too Many A->T Variants
written by Liyf •
-
Genotype Data And Exome Sequencing Data Sensitivity
written by Liyf •
What is the biologic question you are trying to answer? Exome sequencing will actually give much less data about variants that a 1M snp array, but the 1M snp array may not give you the information you want.
as far as I know after finding a significant variants by exome sequencing you should confirm it by sanger sequencing method.On the basis of read number you can decide about the sequence related to specific region if read number is good you should confirm by sanger sequencing.If read number is not enough it might be false result or coverage in that region is not good.