In fboehm/reducedscan: QTL Scans with Simpler Statistical Models in Multiparental Populations

(Last modified: r params$lastmod.)

Overview

I want to adapt the MCMC from qtlbim by adding steps that infer the allelic series, ie, the configuration and the number of alleles.

One approach is to borrow from @jannink2003estimating by updating the allelic number and, in a separate step the configuration for a fixed allelic number.

I may ultimately want to write this in C++, but, for now, I'll write in R.

Reproduce the qtlbim mcmc in R code

I'll first attempt to write the qtlbim mcmc algorithm in R. That way, I will gain understanding of the algorithm while having a R coding of it. I'll also be able to confirm that I've coded it correctly, since I can check results against those of Brian Yandell's qtlbim.

Deriving the MCMC algorithm for qtlbim

We first rederive the algorithm for the case of two founders, as in qtlbim software.

I'm interested in the case of a single trait and, potentially, multiple QTL for that one trait.

Start with the joint likelihood for observed and unobserved quantities.

References

fboehm/reducedscan documentation built on April 30, 2021, 8:31 p.m.