# Downstream data analysis
# Laod library
# 
# 

require("ggplot2")
require("ComplexHeatmap")
#require("ggplot")

th0th9 <- read.table(file = "~/Dropbox/suyasha/Th0_th9.txt",header = T)
th9_aregko <- read.table(file = "~/Dropbox/suyasha/Th9_Th9_aregko.txt",header = T)
sel_genes_values <- read.table(file = "~/Dropbox/suyasha/selected_genes_with_values.txt",header = T,row.names = 1)
colnames(sel_genes_values) <- c("Th9 WT","Th9 Areg-/-")

########
cmn_genes <- intersect(th0th9$gene,th9_aregko$gene)
th0th9_cmn <- th0th9[th0th9$gene %in% cmn_genes,]
th9aregko_cmn <- th9_aregko[th9_aregko$gene %in% cmn_genes,]
rownames(th0th9_cmn) <- th0th9_cmn$gene
rownames(th9aregko_cmn) <- th9aregko_cmn$gene

####
thoth9_lfc <- th0th9_cmn[cmn_genes,"log2.fold_change."]
th9aregko_lfc <- th9aregko_cmn[cmn_genes,"log2.fold_change."]

lfc_data <- cbind(thoth9_lfc,th9aregko_lfc)
colnames(lfc_data) <- c("Th9 WT","Th9 Areg-/-")

library(circlize)
col_fun = colorRamp2(c(-1, 0, 1), c("green", "white", "red"))
col_fun(seq(-3, 3))

Heatmap(as.matrix(lfc_data),cluster_columns =F,name = "LFC",col = col_fun,column_names_rot = 0,
        column_names_centered = T, column_names_gp = gpar(fontsize=14))

Heatmap(as.matrix(sel_genes_values),cluster_columns =F,name = "LFC",col = col_fun,column_names_rot = 0,
        column_names_centered = T, column_names_gp = gpar(fontsize=14),
        row_names_centered = F, row_names_gp = gpar(fontsize=14))


#############################################################################################
setwd("~/Dropbox/suyasha/remetabolitesheatmap")
cell_global <- read.table(file = "~/Dropbox/suyasha/remetabolitesheatmap/Cell_extract_global.txt",header = T,sep = "\t")
cell_wt <- cell_global[cell_global$Condition == "Th9-WT",]
cell_th9 <- cell_global[cell_global$Condition == "Th9-HIF1? KO",]
#cell_data <- cbind(cell_wt,cell_th9)
cell_data <- cbind(cell_th9, cell_wt)
cell_data_gene_name <- cell_data$Name
cell_data_hm <- cell_data[,c(3:5,8:10)]
rownames(cell_data_hm) <- cell_data_gene_name
# Changed the oriantation th9 was first initiallt
colnames(cell_data_hm) <- c("Th9-HIF1α.1","Th9-HIF1α.2","Th9-HIF1α.3","Th9-WT.1","Th9-WT.2","Th9-WT.3")

#xn <- c("5-Oxoproline","5-Oxoproline","Ac-choline","Ac-choline","Aconitate","Aconitate","AMP","AMP","AMP/ATP","AMP/ATP","Arg","Arg","Carbamoyl-Asp","Carbamoyl-Asp","Cit","Cit","Creatine/P-Creatine","Creatine/P-Creatine","Creatinine","Creatinine","Cys","Cys","Cytosine","Cytosine","dA","dA","dC","dC","DMG","DMG","dU","dU","Folate","Folate","Fru","Fru","Glc","Glc","GlcN","GlcN","GlcNAc","GlcNAc","GlucA","GlucA","Glycerol","Glycerol","GSH","GSH","GSH/GSSG","GSH/GSSG","His","His","Homocysteine","Homocysteine","HS2O3","HS2O3","Hypoxanthine","Hypoxanthine","Leu/Ile","Leu/Ile","Lys","Lys","Met","Met","Nicotinamide","Nicotinamide","Ornithine","Ornithine","P-Creatine","P-Creatine","P-EtA","P-EtA","Palmitate","Palmitate","PEP","PEP","Phe","Phe","Pipecolate","Pipecolate","PO4","PO4","PPi","PPi","Pyr","Pyr","Rib","Rib","Sorbitol","Sorbitol","Succ","Succ","T","T","Trp","Trp","Tyr","Tyr","UDP-Glc","UDP-Glc","Uracil","Uracil","Val","Val","Vit-C","Vit-C","Xanthine","Xanthine")
xm <- c("13BPG","13BPG","2-HG","2-HG","2-Oxoadipate","2-Oxoadipate","2-Oxosuccinamate","2-Oxosuccinamate","3PG","3PG","4-OH-PheLac","4-OH-PheLac","5M-adenosine","5M-adenosine","a-KG","a-KG","Ac-carnitine","Ac-carnitine","Adenine","Adenine","ADP","ADP","ADP/ATP","ADP/ATP","Ala","Ala","Arg-Succ","Arg-Succ","Asn","Asn","Asp","Asp","ATP","ATP","C","C","Carnitine","Carnitine","CDP","CDP","CDP-EtA","CDP-EtA","Choline","Choline","Citrulline","Citrulline","CMP","CMP","CoA","CoA","Creatine","Creatine","CTP","CTP","Cys Sulfinic Acid","Cys Sulfinic Acid","Cystathionine","Cystathionine","dG","dG","DHAP","DHAP","F16BP","F16BP","Fum","Fum","G","G","G6P-F6P","G6P-F6P","GABA","GABA","GDP","GDP","GlcA","GlcA","GlcN-6P","GlcN-6P","GlcNAc-6P","GlcNAc-6P","Gln","Gln","Glu","Glu","Gly","Gly","Glycerol-3-P","Glycerol-3-P","GMP","GMP","GSSG","GSSG","IMP","IMP","Inosine","Inosine","Inositol","Inositol","Isobutyryl-CoA","Isobutyryl-CoA","Lac","Lac","Mal","Mal","NAD+","NAD+","NADH","NADH","NADP+","NADP+","Orotate","Orotate","Orotidine","Orotidine","P-Choline","P-Choline","P-Ser","P-Ser","Pantothenate","Pantothenate","Pro","Pro","Pro-OH","Pro-OH","R5P","R5P","S7P","S7P","SAM","SAM","Sarc","Sarc","Ser","Ser","Thr","Thr","Thymine","Thymine","Tyramine","Tyramine","UDP","UDP","UDP-GlcNAc","UDP-GlcNAc","UMP","UMP")
cell_data_hm <- cell_data_hm[unique(xm),]
n <- t(apply(cell_data_hm, 1, function(x)(x-min(x))/(max(x)-min(x))))
n_scaled <- scale(n)
n_scaled_NA <- na.omit(n_scaled)

# color function
col_func = colorRamp2(c(-2, 0, 2), c("blue", "white", "red"))
col_func(seq(-3, 3))

dev.off()
#Heatmap(as.matrix(n_scaled_NA),cluster_columns =F,column_title = "Global", col = col_func,
#        name = "exp",show_row_names = T, show_column_names = T,column_dend_reorder = F,column_names_rot = 0,
#        column_names_centered = T, column_names_gp = gpar(fontsize=14))

Heatmap(as.matrix(n_scaled_NA),cluster_columns =T,column_title = "Global", col = col_func,
        name = "exp",show_row_names = T, show_column_names = T,column_dend_reorder = F,
        column_names_centered = T, row_names_gp = gpar(fontsize=12,fontface="bold"),column_km = 2,border = T)


cell_data_hm_scaled <- n_scaled_NA

Glycolytic_Metabolites <- c("13-BPG","3PG","Ala","DHAP","F16BP","Fru","G6P-F6P","Glc","Gly","Lac","P-Ser","PEP","Pyr","Ser","UDP-Glc")
Glyco <- intersect(rownames(cell_data_hm_scaled),Glycolytic_Metabolites)
dev.off()
Heatmap(as.matrix(cell_data_hm_scaled[Glyco,]),cluster_columns =F,column_title = "Glycolytic Metabolites",
        name = "exp",show_row_names = T, show_column_names = T,column_dend_reorder = F,
        column_names_centered = T, row_names_gp = gpar(fontsize=12,fontface="bold"),border = T)

TCA_Metabolites <- c("a-KG","Aconitate","Asn","Asp","Cit","Fum","Gln","Glu","Lac","Mal","Palmitate","Pyr","Succ")
TCA <- intersect(rownames(cell_data_hm_scaled),TCA_Metabolites)
dev.off()
Heatmap(as.matrix(cell_data_hm_scaled[TCA,]),cluster_columns =F,column_title = "TCA Metabolites",
        name = "exp",show_row_names = T, show_column_names = T,column_dend_reorder = F,
        column_names_centered = T, row_names_gp = gpar(fontsize=12,fontface="bold"),border = T)

PPP_Metabolites <- c("G6P-F6P","Glc","GlcA","IMP","Lac","R5P","Rib","S7P","UMP")
PPP <- intersect(rownames(cell_data_hm_scaled),PPP_Metabolites)
dev.off()
Heatmap(as.matrix(cell_data_hm_scaled[PPP,]),cluster_columns =F,column_title = "PPP Metabolites",
        name = "exp",show_row_names = T, show_column_names = T,column_dend_reorder = F,
        column_names_centered = T, row_names_gp = gpar(fontsize=12,fontface="bold"),border = T)

Currency_Metabolites <- c("ADP","ADP/ATP","AMP","AMP/ATP","ATP","Creatine","Creatine/P-Creatine","Creatinine","GSH","GSH/GSSG","GSSG","NAD+","NADH","NADP+","P-Creatine")
Currency <- intersect(rownames(cell_data_hm_scaled),Currency_Metabolites)
dev.off()
Heatmap(as.matrix(cell_data_hm_scaled[Currency,]),cluster_columns =F,column_title = "Currency Metabolites",
        name = "exp",show_row_names = T, show_column_names = T,column_dend_reorder = F,
        column_names_centered = T, row_names_gp = gpar(fontsize=12,fontface="bold"),border = T)

Amino_acids <- c("2-HG","Ala","Arg","Asn","Asp","Cys","Gln","Glu","Gly","His","Leu/Ile","Lys","Met","Phe","Pro","Ser","Thr","Trp","Tyr","Val")
Amino <- intersect(rownames(cell_data_hm_scaled),Amino_acids)
dev.off()
Heatmap(as.matrix(cell_data_hm_scaled[Amino,]),cluster_columns =F,column_title = "Amino acids",
        name = "exp",show_row_names = T, show_column_names = T,column_dend_reorder = F,
        column_names_centered = T, row_names_gp = gpar(fontsize=12,fontface="bold"),border = T)

Fatty_Acids_Intermediates  <- c("Ac-carnitine","Ac-choline","CDP-EtA","Choline","Glycerol-3-P","Glycerol","P-Choline","P-EtA","Palmitate")
Fatty <- intersect(rownames(cell_data_hm_scaled),Fatty_Acids_Intermediates)
dev.off()
Heatmap(as.matrix(cell_data_hm_scaled[Fatty,]),cluster_columns =F,column_title = "Fatty Acids Intermediates",
        name = "exp",show_row_names = T, show_column_names = T,column_dend_reorder = F,
        column_names_centered = T, row_names_gp = gpar(fontsize=12,fontface="bold"),border = T)

footprint_data <- read.table(file = "~/Dropbox/suyasha/remetabolitesheatmap/Footprint_data.txt",header = T,sep = "\t")
foot_wt <- footprint_data[footprint_data$Condition == "Th9-WT",]
foot_th9 <- footprint_data[footprint_data$Condition == "Th9-HIF1? KO",]
#foot_data <- cbind(foot_wt,foot_th9)
foot_data <- cbind(foot_th9,foot_wt)
#foot_data[is.na(foot_data)] <- 0
foot_data_gene_name <- foot_data$Name
foot_data_hm <- foot_data[,c(3:5,8:10)]
#rownames(foot_data_hm) <- foot_data_gene_name
colnames(foot_data_hm) <- c("Exp1.ko","Exp2.ko","Exp3.ko", "Exp1.wt","Exp2wt","Exp3wt")
rownames(foot_data_hm) <- foot_data_gene_name
foot_genes <- c("Ala","Ala","Cys","Cys","dC","dC","Glc","Glc","Glu","Glu","Gly","Gly","Lac","Lac","Lys","Lys","Met","Met","Phe","Phe","Ser","Ser","Trp","Trp","Tyr","Tyr")
foot_data_hm <- foot_data_hm[unique(foot_genes),]
foot_data_hm <- na.omit(foot_data_hm)
m <- t(apply(foot_data_hm, 1, function(x)(x-min(x))/(max(x)-min(x))))
foot_data_hm_scaled <- scale(m)
#foot_data_hm_scaled <- scale(foot_data_hm)
Heatmap(as.matrix(foot_data_hm_scaled),cluster_columns =F,name = "exp",show_row_names = T, show_column_names = T,column_title = "Footprinting",
        column_names_centered = T, row_names_gp = gpar(fontsize=12,fontface="bold"),border = T)
#cluster_columns = cluster_within_group(mat, group)

################### PCA ########################################################
cell_pca_data <- read.table(file = "~/Dropbox/suyasha/remetabolitesheatmap/Cell extract(Suyasha)-PCA.csv",header = T,sep = ",")
colors <- c("red","red","red", "green","green","green")
scatterplot3d(x = cell_pca_data$PC1,y = cell_pca_data$PC2,z = cell_pca_data$PC3, 
              xlab='PCA.1', ylab='PCA.2', zlab='PCA.3', color = colors, pch=16,
              main = "Principal Component Analysis")
legend("topright", legend = c("Th9 Hif1αkd","Th9 WT"),
       col = c("red", "green"), pch = 16)