Determining Optimal Number of Samples to include in Panel of NormalAnswered
We have been following tutorial(s) 11682 and 11683 for sensitive detection of copy ratio alterations in benign tissue utilizing the PON sensitivity in place of a matched normal when one may not be easily accessible. Currently, we have a PON of 47 germlines generated by "CreateReadCountPanelOfNormals".
Although I have read in many posts that a range between 40-100 specimen is good. How and/or is there a way to determine the optimum number of samples to include in the PON? We have the ability to potentially add several hundred to our panel but are not sure at what point this becomes excessive (both financially and temporally).
Thank you in advance for your help
I am going to move your post into our Community Discussions -> Documentation Questions topic, as the Germline topic is for reporting bugs and issues with GATK.
You can read more about our forum guidelines and the topics here: Forum Guidelines.
Hi Gannon Cottone,
There isn't a method as of now to find an optimal number of samples for the PON including the associated costs.
It's not the same type of Panel of Normals, but one of our developers answered a somewhat similar question regarding the Mutect2 PON during the Mutect2 webinar: https://gatk.broadinstitute.org/hc/en-us/community/posts/360074659231-What-s-the-minimal-number-of-normal-samples-required-to-generate-the-SomaticPanelOfNormals-
The more the better was the recommendation, but if the costs are a concern, 40 is also great and has many benefits over not using a panel of normals.
Please sign in to leave a comment.