reworkhow / JWAS.jl

Julia for Whole-genome Analysis Software
http://QTL.rocks
GNU General Public License v2.0
96 stars 44 forks source link

Pi for multi-trait analyisis under Bayes B method #76

Closed bosamu05 closed 3 years ago

bosamu05 commented 3 years ago

Hello dear PhDs

My queries are about the pi (π) values for multi-trait Bayes B. Which are standard values for a trivariate analysis? How can I change those values when our a priori hypothesis is that most SNPs have zero effect? What is the argument to be changed? In a nutshell, Doing a multi-trait analysis without estimating pi. For example. pi_1: 0.95, pi_2=0.95, pi_3 =0.95; Pr(0,0,0) = probbaility that a given SNP has zerro effect for three traits = 0.857. The rest of prrobabilty variables are from a uniform distribution).

Thank you in advances