MRPC (version 2.2.0)

SimulateData: Simulate data under certain graphs

Description

This function simulates data using linear models for several graphs: the five basic topologies and three topologies that are common in biology, namely the multi-parent graph, the star graph and the layered graph. See references for details.

Usage

SimulateData(N, p, model, b0.1, b1.1, b1.2, b1.3, sd.1)

Arguments

N

The number of observations.

p

Population frequency of the reference allele. Real number between 0 to 1, which is the number of a particular allele is present.

model

The model for which data will be simulated. For example, if you want to generate data for model 0 you would type 'model0' into the function.

b0.1

Intercept of b0.1 + b1.1*P1 + b1.2*P2 + b1.3*P3, where P1, P2, and P3 are the parents of the corresponding node.

b1.1

Slope of P1 for b0.1 + b1.1*P1 + b1.2*P2 + b1.3*P3, where P1, P2, and P3 are the parents of the corresponding node.

b1.2

Slope of P2 for b0.1 + b1.1*P1 + b1.2*P2 + b1.3*P3, where P1, P2, and P3 are the parents of the corresponding node.

b1.3

Slope of P3 for b0.1 + b1.1*P1 + b1.2*P2 + b1.3*P3, where P1, P2, and P3 are the parents of the corresponding node.

sd.1

Standard deviation for corresponding data generated nodes.

Value

Matrix

Details

The first column of the input matrix is the genotype of the expression quantitative trait loci (eQTL)/Copy number variation (CNVs) and the remaining columns are the node expression data.

References

1. Badsha MB and Fu AQ (2019). Learning causal biological networks with the principle of Mendelian randomization. Frontiers in Genetics, 10(460).

2. Badsha MB, Martin EA and Fu AQ (2018). MRPC: An R package for accurate inference of causal graphs. arXiv:1806.01899.

See Also

MRPC; SimulateDataNP, which simulates data for a node with no parent; SimulateData1P for a node with one parent; SimulateData2P for a node with two parents.

Examples

# NOT RUN {
# Data pre-processing

# If you use only one genotype of the expression quantitative trait loci
# (eQTL)/Copy number variation (CNV), the 1st column of
# the input matrix will be #eQTL/CNV and the remaining
# columns are the gene expression data.


## Model 0
simu_data_M0 <- SimulateData(N = 10^3, 
                             p = 0.45,
                             'model0', 
                             b0.1 = 0,
                             b1.1 = 1, 
                             b1.2 = 1,
                             b1.3 = 1, 
                             sd.1 = 1)
                             
## Model 1
simu_data_M1 <- SimulateData(N = 10^3, 
                             p = 0.45,
                             'model1', 
                             b0.1 = 0,
                             b1.1 = 1, 
                             b1.2 = 1,
                             b1.3 = 1, 
                             sd.1 = 1)
                             
## Model 2
simu_data_M2 <- SimulateData(N = 10^3, 
                             p = 0.45,
                             'model2', 
                             b0.1 = 0,
                             b1.1 = 1, 
                             b1.2 = 1,
                             b1.3 = 1, 
                             sd.1 = 1)
                             
## Model 3
simu_data_M3 <- SimulateData(N = 10^3, 
                             p = 0.45,
                             'model3', 
                             b0.1 = 0,
                             b1.1 = 1, 
                             b1.2 = 1,
                             b1.3 = 1, 
                             sd.1 = 1)
                             
## Model 4
simu_data_M4 <- SimulateData(N = 10^3, 
                             p = 0.45,
                             'model4', 
                             b0.1 = 0,
                             b1.1 = 1, 
                             b1.2 = 1,
                             b1.3 = 1, 
                             sd.1 = 1)
                             
## Multiple Parent Model
simu_data_multiparent <- SimulateData(N = 10^3, 
                                      p = 0.45,
                                      'multiparent', 
                                      b0.1 = 0,
                                      b1.1 = 1, 
                                      b1.2 = 1,
                                      b1.3 = 1, 
                                      sd.1 = 1)

## Star Model
simu_data_starshaped <- SimulateData(N = 10^3, 
                                     p = 0.45,
                                     'starshaped', 
                                     b0.1 = 0,
                                     b1.1 = 1, 
                                     b1.2 = 1,
                                     b1.3 = 1, 
                                     sd.1 = 1)

## Layered Model
simu_data_layered <- SimulateData(N = 10^3, 
                                  p = 0.45,
                                  'layered', 
                                  b0.1 = 0,
                                  b1.1 = 1, 
                                  b1.2 = 1,
                                  b1.3 = 1, 
                                  sd.1 = 1)
# }