plot_interaction_model: Plot interaction model results

In MethReg: Assessing the regulatory potential of DNA methylation regions or sites on gene transcription

Description

Create several plots to show interaction data TF expression with target gene interaction using a linear model

log2(RNA target) = log2(TF) + DNAm + log2(TF) * DNAm

To consider covariates, RNA can also be the residuals.

log2(RNA target residuals) = log2(TF residual) + DNAm + log2(TF residual) * DNAm

Usage

plot_interaction_model(
  triplet.results,
  dnam,
  exp,
  metadata,
  tf.activity.es = NULL,
  tf.dnam.classifier.pval.thld = 0.001,
  label.dnam = "beta-value",
  label.exp = "expression",
  genome = "hg38"
)

Arguments

triplet.results	Output from function interaction_model with Region ID, TF (column name: TF), and target gene (column name: target), p-values and estimates of interaction
dnam	DNA methylation matrix or SummarizedExperiment object (columns: samples same order as met, rows: regions/probes)
exp	gene expression matrix or a SummarizedExperiment object (columns: samples same order as met, rows: genes)
metadata	A data frame with samples as rownames and one columns that will be used to color the samples
tf.activity.es	A matrix with normalized enrichment scores for each TF across all samples to be used in linear models instead of TF gene expression.
tf.dnam.classifier.pval.thld	P-value threshold to consider a linear model significant of not. Default 0.001. This will be used to classify the TF role and DNAm effect.
label.dnam	Used for label text. Option "beta-value" and "residuals"
label.exp	Used for label text. Option "expression" and "residuals"
genome	Genome of reference to be added to the plot as text

Value

A ggplot object, includes a table with results from fitting interaction model, and the the following scatter plots: 1) TF vs DNAm, 2) Target vs DNAm, 3) Target vs TF, 4) Target vs TF for samples in Q1 and Q4 for DNA methylation, 5) Target vs DNAm for samples in Q1 and Q4 for the TF

Examples

library(dplyr)
dnam <- runif(20, min = 0,max = 1) %>% sort %>%
  matrix(ncol = 1) %>%  t
rownames(dnam) <- c("chr3:203727581-203728580")
colnames(dnam) <- paste0("Samples",1:20)

exp.target <-  runif(20,min = 0,max = 10) %>% sort %>%
  matrix(ncol = 1) %>%  t
rownames(exp.target) <- c("ENSG00000232886")
colnames(exp.target) <- paste0("Samples",1:20)

exp.tf <- runif(20,min = 0,max = 10) %>%
  matrix(ncol = 1) %>%  t
rownames(exp.tf) <- c("ENSG00000101412")
colnames(exp.tf) <- paste0("Samples",1:20)

exp <- rbind(exp.tf, exp.target)

triplet <- data.frame(
   "regionID" =  c("chr3:203727581-203728580"),
   "target" = "ENSG00000232886",
   "TF" = "ENSG00000101412"
)

results <- interaction_model(
  triplet = triplet,
  dnam = dnam,
  exp = exp,
  fdr = FALSE,
  filter.correlated.tf.exp.dna = FALSE
)
plots <- plot_interaction_model(
    triplet.results = results,
    dnam = dnam,
    exp = exp
)

MethReg documentation built on Nov. 8, 2020, 8:01 p.m.