FaST-LMM: Factored Spectrally Transformed Linear Mixed Models
Genetic analysis in structured populations used mixed linear models where the variance matrix of the error term is a linear combination of an identity matrix and a positive definite matrix.
The linear model is of the familiar form: $$y = X \beta + e$$
- $y$: phenotype
- $X$: covariates
- $\beta$: fixed effects
- $e$: error term
Further $V(e) = \sigma_G^2 K + \sigma_E^2 I$, where $\sigma_G^2$ is the genetic variance, $\sigma_E^2$ is the environmental variance, $K$ is the kinship matrix, and $I$ is the identity matrix.
The key idea in speeding up computations here is that by rotating the phenotypes by the eigenvectors of $K$ we can transform estimation to a weighted least squares problem.
This implementation is my attempt to learn Julia and numerical linear algebra. The code is being tested.
Guide to the directories:
src
: Julia source codedata
: Example data for development and testingtest
: Code for testingdocs
: Notes on comparisons with other implementations