search

rwdavies / STITCH

STITCH - Sequencing To Imputation Through Constructing Haplotypes
http://www.nature.com/ng/journal/v48/n8/abs/ng.3594.html
GNU General Public License v3.0
77 stars 17 forks source link

How to set K and nGen for plant #36

Closed SC-Duan closed 4 years ago

SC-Duan commented 4 years ago

Hi, I want to impute a plant in zingiberaceae using STITCH, and there are not any information about K and Ne, how can I determine (test) the K and nGen? I have 1500 samples and the genome size is ~3.2Gb, test different K may waste more time, is there some experience for set K and Ne? Thank you!

rwdavies commented 4 years ago

I would recommend having a read through this https://github.com/rwdavies/STITCH#note-on-the-selection-of-k-and-ngen and let me know if you have additional questions? Simplly put, there's not an optimal way to do it, it requires a bit of trial and error for K. nGen should be more straightforward, the method is more robust to that parameter, and you use 4 * Ne / K for some estimate of the effective population size Ne if wild, and if non-wild, you can use some estimate of the number of generations since founding

SC-Duan commented 4 years ago

Thank you!