Sample Size For Tumor Normal Ngs Study

Question

[learnerforever]

In a recent paper http://genesdev.cshlp.org/content/25/6/534.long

It has been quoted that for exome-sequencing "it was calculated that 500 tumor samples are needed to detect, with ∼80% power, genes that are mutated in 3% of patients (assuming a typical background mutation rate)"

The reference cited also does not clarify how the number was derived.

Could someone show exactly how this calculation was done assuming an example mutation rate of say 0.01 (the paper does not say what mutation rate was used) This seems to be a very standard process to assess the number of pairs, but I could not find a definitive example/tutorial showing how this number could be calculated.

More generally, I'd like to know how to calculate the following:

Find N (number of pairs), to detect with power B mutations that are present in P % of samples assuming a mutation rate of M, where M is average mutation rate of a gene based on prior knowledge"

Really appreciate help from this forum!

Answer 1

[Neilfws]

You might find the following links useful:

Article: Statistical Guidance for Experimental Design and Data Analysis of Mutation Detection in Rare Monogenic Mendelian Diseases by Exome Sequencing

Associated website: tools for exome power calculation

Note that entering certain (presumably invalid) values into the power calculator generates a blank page.

[learnerforever]

This was very useful! Is there any similar work/calculator for paired tumor-normal analysis and detecting somatic rather than inherited mutations?