trainLocalModel: Train a local-only model for a phenotype

In bhattacharya-a-bt/MOSTWAS: Multi-Omic Strategies for Transcriptome-Wide Association Studies

Train a local-only model for a phenotype

Description

The function runs FUSION-type local-only modeling for a phenotype

Usage

trainLocalModel(
  phenoInt,
  midSNP,
  mediator,
  medLocs,
  covariates,
  cisDist = 5e+05,
  nfolds = 5,
  seed = 1218
)

Arguments

phenoInt	character, phenotype name
midSNP	snpObject, snp Object in the bigsnpr format
mediator	data frame, mediator intensities
medLocs	data frame, mediator locations
covariates	data frame, covariates in MatrixEQTL format
cisDist	numeric, local window, default .5 Mb
nfolds	integer, number of folds
seed	numeric, random seed

Value

list with Model, predicted values, and CV-R2

bhattacharya-a-bt/MOSTWAS documentation built on April 6, 2023, 12:20 a.m.