Run_LORS: Run_LORS
In jdrhyne2/FastLORS: Joint Modeling for eQTL Mapping in R

Description Usage Arguments Value Examples

Run_LORS is a function used to run either FastLORS or LORS

1
2
3

Run_LORS(Y, X, method = "FastLORS", screening = "LORS-Screening",
  tune_method = "FastLORS", seed = 123, maxiter = 10000,
  eps = 2.2204e-16, tol = 1e-04, cross_valid = TRUE, omega_SOR = 1.999)

`Y`	gene expression matrix
`X`	matrix of SNPs
`method`	chooses with modeling method to run
`screening`	Either "LORS-Screening", "HC-Screening", or "None". The default method, LORS-Screening, is recommended if the number of SNPs is large. HC-Screening of Rhyne et al. (2018) is under development but is included here as an option.
`tune_method`	chooses whether FastLORS should be used for parameter tuning or the original LORS procedure should be used. Default is FastLORS
`seed`	random seed to be used for setting training and validation set. Default is 123.
`maxiter`	maximum number of iterations
`eps`	constant used when checking the convergence. Ensures no division by 0.
`tol`	tolerance level for convergence
`cross_valid`	chooses whether cross-validation should be used in parameter tuning. Default is TRUE.
`omega_SOR`	the value of omega to use if applying successive over-relaxation with FastLORS.

`LORS_Obj or Fast_LORS_Obj`	A list produced from LORS or FastLORS containing (1) B: estimate of the coefficient matrix (2) L: estimate of the matrix of hidden factors (3) mu: estiamte of the vector of intercepts (4) f_val_vec: objective function values and (5) res_vec: relative change in objective function values
`selectedSNPs`	The SNPs selected by the screening method
`screening_time`	The time (in seconds) spent on screening step
`param_time`	The time (in seconds) spent on the parameter tuning step
`model_time`	The time (in seconds) spent on the joint modeling step
`total_time`	The time (in seconds) spent on the screening, parameter tuning, and joint modeling steps
`rho`	The value of rho chosen through parameter tuning
`lambda`	The value of lambda chosen through parameter tuning

##Example

## Generate some data
n <- 20
p <- 50
q <- 30
k <- 4
set.seed(123)
X <- matrix(rbinom(n*p,1,0.5),n,p)
L <- matrix(rnorm(n*k),n,k) %*% t(matrix(rnorm(q*k),q,k))
B <- matrix(0, ncol(X), ncol(L))
activeSNPs <- sort(sample(c(1:nrow(B)), 20))
for(i in 1:length(activeSNPs)){
genes_influenced <- sort(sample(c(1:ncol(B)),5))
B[activeSNPs[i], genes_influenced] <- 2
}
E <- matrix(rnorm(n*q),n,q)
Y <- X %*% B + L + E

## Usage
Run_LORS(Y, X, method = "FastLORS")