RunPairwisePrediction: Given each significant pairwise model and the input data,...

In ncats/MultiOmicsGraphPrediction: Network-Based Ensemble Prediction from IntLIM Predictors

Given each significant pairwise model and the input data, predict the phenotype for each sample. Recall that IntLIM models take the following form, where a_i and a_j are a pair of analytes. a_i ~ beta0 + beta1(a_j) + beta2(phenotype) + beta3(a_j:phenotype) + beta4...n(covariates) Therefore, to predict phenotype given the betas learned by IntLIM, we use the following model: p ~ (a_i - (beta0 + beta1(a_j) + beta4...n(covariates)) / (beta2 + beta3(a_j))

Description

Given each significant pairwise model and the input data, predict the phenotype for each sample. Recall that IntLIM models take the following form, where a_i and a_j are a pair of analytes. a_i ~ beta0 + beta1(a_j) + beta2(phenotype) + beta3(a_j:phenotype) + beta4...n(covariates) Therefore, to predict phenotype given the betas learned by IntLIM, we use the following model: p ~ (a_i - (beta0 + beta1(a_j) + beta4...n(covariates)) / (beta2 + beta3(a_j))

Usage

RunPairwisePrediction(
  inputResults,
  inputData,
  stype = "",
  covar = c(),
  independentVarType = 2,
  outcomeType = 1
)

Arguments

inputResults	The data frame of filtered results from IntLIM model and processing results (output of ProcessResults()). All results must include learned covariate weights (i.e. must be run with save.covar.pvals = TRUE)
inputData	An IntLimData object that includes the input.
stype	The phenotype (outcome) to predict. This can be either a categorical or numeric outcome.
covar	The clinical covariates to include in the model. These should be the same covariates that were included when running the IntLIM linear models.
independentVarType	The independent variable type (1 or 2)
outcomeType	The outcome type (1 or 2)

Value

A data frame of predictions, where each column is a sample and each row is a predictor.

ncats/MultiOmicsGraphPrediction documentation built on Aug. 23, 2023, 9:19 a.m.