Added correction for first questions of TP2

Practical2/Correction/Q1to4.R

@@ -0,0 +1,62 @@
+library(tidyverse)
+library(MOFA2)
+# setup reticulate with correct python install
+library(reticulate)
+use_python(system("which python3.7",intern=T), required=TRUE)
+
+# Q1
+#a) read data
+Data.Clin = read_tsv("medical_genomics_course/Practical2/Data/Data.Clin.txt")
+RNA = read.table("medical_genomics_course/Practical2/Data/RNA.tsv",h=T,row.names=1)
+DNAMeth_promoter = read.table("medical_genomics_course/Practical2/Data/DNAMeth_promoter.tsv",h=T,row.names=1)
+DNAMeth_genebody = read.table("medical_genomics_course/Practical2/Data/DNAMeth_genebody.tsv",h=T,row.names=1)
+DNAMeth_enhancer = read.table("medical_genomics_course/Practical2/Data/DNAMeth_enhancer.tsv",h=T,row.names=1)
+
+#b) 
+head(Data.Clin)
+head(RNA)
+dim(RNA)
+head(DNAMeth_promoter)
+dim(DNAMeth_promoter)
+head(DNAMeth_genebody)
+dim(DNAMeth_genebody)
+head(DNAMeth_enhancer)
+dim(DNAMeth_enhancer)
+
+#Q2
+##a) densities
+ggplot( data=tibble(value=unlist(RNA)), mapping = aes(x=value) ) + geom_histogram()
+ggplot( data=tibble(value=unlist(DNAMeth_enhancer)), mapping = aes(x=value) ) + geom_histogram()
+ggplot( data=tibble(value=unlist(DNAMeth_genebody)), mapping = aes(x=value) ) + geom_histogram()
+ggplot( data=tibble(value=unlist(DNAMeth_promoter)), mapping = aes(x=value) ) + geom_histogram()
+### not really Gaussian but not too widespread and globally symmetric
+##b) create mofa object
+mofa_untrained = create_mofa( list(RNA=as.matrix(RNA),Meth_pro=as.matrix(DNAMeth_promoter),Meth_bod=as.matrix(DNAMeth_genebody),Meth_enh=as.matrix(DNAMeth_enhancer)) )
+
+#Q3
+##a)
+opt_train = get_default_training_options(mofa_untrained)
+opt_train$convergence_mode ="slow"
+opt_model = get_default_model_options(mofa_untrained)
+opt_model$num_factors = 5
+
+#b)
+mofa_untrained = prepare_mofa(mofa_untrained,model_options = opt_model,training_options = opt_train)
+
+#c) 
+mofa_trained = run_mofa(mofa_untrained)
+
+#Q4
+##a) 
+plot_variance_explained(mofa_trained)
+
+##b)
+plot_factors(mofa_trained)
+
+##c) 
+plot_factor_cor(mofa_trained)
+
+## clean session
+rm(mofa_untrained)
+rm(list = c("DNAMeth_enhancer",  "DNAMeth_genebody",  "DNAMeth_promoter","RNA","file","opt_model","opt_train"))
+gc()