/
enrichment_ternaryplot.R
246 lines (201 loc) · 12.7 KB
/
enrichment_ternaryplot.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
### CREATE TERNARY PLOTS BASED ON KO COUNTS & KO ENRICHMENT ANALYSIS, BETWEEN COMPARTMENTS AND PLANT SPECIES ###
# SETTINGS #
# WORKING DIRECTORIES
setwd("/my_path")
data.dir=c("/my_data") #metagenomic data
main.dir=c("/my_path/main") #enrichment analysis results
result.dir=c("/my_path/result") #saved results
dir.create(result.dir, recursive = T, showWarnings = F)
# LIBRARIES
library(data.table)
library(ggplot2)
library(ggrepel)
library(Rmisc)
library(tidyr)
library(ggtern)
library(extrafont) #font_import() loadfonts(device = "win")
# PLOT SETTINGS
tema=theme(axis.text = element_text(color="black",size=11, angle=0, family="Times New Roman"),
axis.title = element_text(color="black",size=11, family="Times New Roman"),
legend.text = element_text(color = "black",size=11, family="Times New Roman"),
panel.border =element_rect(color = "black", fill = NA) , #element_blank(),
panel.background = element_rect(fill = "white",colour = "white",size = 0.5, linetype = "solid"),
panel.grid.major.y = element_blank(),panel.grid.minor.x = element_blank(),
legend.position = "left") #tema
# LOAD METADATA
path.data=fread(file = paste(getwd(),"kegg_pathway_data.txt",sep="/"),sep="\t", header = T)
pro.data=read.delim(paste(getwd(),"kegg_process_pathway_data.txt",sep = "/"), quote = "", stringsAsFactors = F)
full.meta=read.delim(paste(data.dir,"meta_data.txt",sep = "/"),header = T,stringsAsFactors = F)
# GET FILE NAMES
files=grep(33000,list.files(data.dir),value = T) #id of samples
keys=full.meta$Genome_name ; names(keys)=full.meta$taxon_oid #match id and sample names
# LOAD COUNTS
load(file=paste(getwd(),"counts_per_taxa_domain","taxon.ko.counts.list",sep = "/"))
data.full=taxon.tab.list[["Bacteria"]]+taxon.tab.list[["Archaea"]]
# ORGANIZE DESIRED PATHWAYS
phototrophy=c("Porphyrin and chlorophyll metabolism [PATH:ko00860]","Carotenoid biosynthesis [PATH:ko00906]","Photosynthesis [PATH:ko00195]")
phosp.ox=c("Oxidative phosphorylation [PATH:ko00190]")
biofilm=c("Quorum sensing [PATH:ko02024]","Biofilm formation - Pseudomonas aeruginosa [PATH:ko02025]","Biofilm formation - Escherichia coli [PATH:ko02026]", "Flagellar assembly [PATH:ko02040]","Biofilm formation - Vibrio cholerae [PATH:ko05111]","Bacterial chemotaxis [PATH:ko02030]")
carbon.met=c("Methane metabolism [PATH:ko00680]","Glyoxylate and dicarboxylate metabolism [PATH:ko00630]","Propanoate metabolism [PATH:ko00640]","Butanoate metabolism [PATH:ko00650]")
c.fixation=c("Carbon fixation pathways in prokaryotes [PATH:ko00720]","Carbon fixation in photosynthetic organisms [PATH:ko00710]")
transporters=c("ABC transporters [PATH:ko02010]","Phosphotransferase system (PTS) [PATH:ko02060]")
two.component=c("Two-component system [PATH:ko02020]")
nitrogen.met=c("Nitrogen metabolism [PATH:ko00910]")
aminoacid=pro.data[pro.data$process=="Amino acid metabolism","pathway"]
xenobiotics=pro.data[pro.data$process=="Xenobiotics biodegradation and metabolism","pathway"]
sugar.met=c("Fructose and mannose metabolism [PATH:ko00051]","Starch and sucrose metabolism [PATH:ko00500]")
natural.products=unique(path.data[c(5518:7032),pathway])[-5]
sulfur.met=c("Sulfur metabolism [PATH:ko00920]")
#EXTRACT THE KO IDS OF DESIRED PATHWAYS
process=list(
Phototrophy=path.data[path.data$pathway %in% phototrophy | path.data$anygene_name %in% c("pufM","pufL","pufA","pufB","pufX","puhA","pucA","pucB"),ko_id],
Oxidative_phosphorilation=path.data[path.data$pathway %in% phosp.ox,ko_id],
Biofilm_QS=path.data[path.data$pathway %in% biofilm,ko_id],
Natural_products=path.data[path.data$pathway %in% natural.products,ko_id],
Xenobiotic_metabolism=path.data[path.data$pathway %in% xenobiotics & !(path.data$anygene_name %in% c("nifD", "nifK","nifH","anfG","vnfD","vnfK", "vnfG","vnfH")),ko_id ],
Nitrogen_fixation=path.data[path.data$anygene_name %in% c("nifD", "nifK","nifH","anfG","vnfD","vnfK", "vnfG","vnfH") , ko_id],
Carbon_metabolism=path.data[path.data$pathway %in% carbon.met,ko_id],
Sugar_metabolism=path.data[path.data$pathway %in% sugar.met,ko_id],
Carbon_fixation=path.data[path.data$pathway %in% c.fixation,ko_id],
Nitrogen_Aminoacids=path.data[path.data$pathway %in% nitrogen.met | path.data$pathway %in% aminoacid,ko_id],
PTS_ABC_transportes=path.data[path.data$pathway %in% transporters,ko_id],
Two_component=path.data[path.data$pathway %in% two.component & !(path.data$anygene_name %in% c("pufM","pufL","pufA","pufB","pufX","puhA","pucA","pucB")) , ko_id]
#sulfur.metabolism=path.data[path.data$pathway %in% sulfur.met,"ko_id"]
) #process list
#SELECT YOUR COLOR PALETTE
my_aes3=c("#024AAC","#168582","#2DBADA","#358404","#F48F06","#F2252C","#E2E41C","#10B62F","#95C331","#A30AB0", "#E283C4","hotpink")
getPalette=colorRampPalette(my_aes3)
aes3=data.frame(color=getPalette(length(names(process))), ruta = names(process)) #color assignment
for (h in 1:2){
# PLOT BETWEEN COMPARTMENTS #
# TRANSFORM COUNTS
result_norm=as.data.frame(t(t(data.full)/colSums(data.full))) #relative abundance
result_norm$ko_id=rownames(result_norm) #ko names
result.tidy=gather(result_norm, key = sample, value = rel.abundance, keys[files]) #create tidy data
for (i in keys){
result.tidy[result.tidy$sample==i,c(4,5,6)]=full.meta[full.meta$Genome_name==i,c("species","site","compartment")]
} #add metadata
# SELECT DATA
file=c("Expanded_families_edgeR.txt","Reduced_families_edgeR.txt","Comparative_Genomics_families_edgeR.txt")
comparison=c("SYM_Prokaryotes_ko_75", "TEQ_Prokaryotes_ko_0.75")[h] #select the comparison
if (sum(grep("SYM",comparison))>0){
pair=c("phyllo_vs_rhizo","phyllo_vs_soil", "rhizo_vs_soil")
datos=result.tidy[result.tidy$species != "A.tequilana" & !(result.tidy$site %in% c("Penjamo","Amatitan")),]
} else if (sum(grep("TEQ",comparison)>0)){
pair=c("phyllo_vs_rhizo","phyllo_vs_soil", "rhizo_vs_soil")
datos=result.tidy[result.tidy$species == "A.tequilana" | result.tidy$site %in% c("Penjamo","Amatitan"),]
} #select tha data from the tidy table based on the comparison
sub1.1=read.delim(paste(main.dir,comparison,pair[1],file[1], sep = "/"))
sub1.2=read.delim(paste(main.dir,comparison,pair[1],file[2], sep = "/"))
sub2.1=read.delim(paste(main.dir,comparison,pair[2],file[1], sep = "/"))
sub2.2=read.delim(paste(main.dir,comparison,pair[2],file[2], sep = "/"))
sub3.1=read.delim(paste(main.dir,comparison,pair[3],file[1], sep = "/"))
sub3.2=read.delim(paste(main.dir,comparison,pair[3],file[2], sep = "/")) #read the expanded[1] and reduced[2] KOs per comparison [1,2,3]
# TRANSFORM DATA
mean.data=summarySE(data=datos, measurevar = "rel.abundance", groupvars = c("ko_id","compartment")) #calculate the mean relative abundances between compartmens
mean.data2=mean.data[,c(1,2,4)] #select necessary data
plot.data=spread(mean.data2, key = compartment, value = rel.abundance) #transform data from tidy to untidy in order to have the 3 axis for the triplot
# SUBSET ONLY COMPARED KOs
all=read.delim(paste(main.dir,comparison,pair[1],file[3], sep = "/")) #read all the compared KOs
plot.data=plot.data[plot.data$ko_id %in% rownames(all),] #select them from data
# LABEL SIGNIFICANT DATA
plot.data$value="No significant"
plot.data[plot.data$ko_id %in% c(rownames(sub1.1),rownames(sub1.2),rownames(sub2.1),rownames(sub2.2),rownames(sub3.1),rownames(sub3.2)),"value"]="Significant"
# HIGHLIGHT PROCESSES
sub.process=process[names(process)[c(1:6)]]
plot.data$ruta="Other"
for (i in rev(names(sub.process))){
plot.data[plot.data$ko_id %in% sub.process[[i]], "ruta"]=i
} #label each KO with a process
plot.data$ruta=factor(plot.data$ruta, levels = c("Other",names(sub.process)))
# ADD FDR
all=all[plot.data$ko_id,]
plot.data$p.value=all$FDR #non significant FDR will be based on the first comparison pair[1]
p=rbind(sub1.1,sub1.2,sub2.1,sub2.2,sub3.1,sub3.2)
for (i in rownames(p)){
plot.data[plot.data$ko_id==i,"p.value"]=p[i,"FDR"]
} #subtitute the FDR value of significan data of each comparison
# CREATE CHEAT DATSET TO HIGHLIGHT SPECIFIC POINTS
cheat=plot.data[plot.data$ruta %in% names(sub.process) & plot.data$value == "Significant",]
# PLOT
a=ggtern(data=plot.data, aes(x=phyllosphere, y=rhizosphere, z=soil,shape=value, color=ruta, size=-log10(p.value)))+
geom_mask()+
geom_point()+
geom_point(data =cheat, aes(x=phyllosphere, y=rhizosphere, z=soil,shape=value, color=ruta, size=-log10(p.value)))+
scale_shape_manual(values=c(21,19))+
scale_size_continuous(range = c(1,2))+
scale_color_manual(values =c("grey70",as.character(aes3[aes3$ruta %in% names(sub.process),1])))+
ylab("")+xlab("")+zlab("")+tema
#SAVE
png(filename = paste(result.dir,paste(comparison,"ternary_9","png",sep = "."),sep = "/"),
width = 160, height = 100, units = "mm", pointsize = 15, bg = "white", res=300, type="cairo")
print(a) ; dev.off()
# PLOT BETWEEN SPECIES #
# TRANSFORM COUNTS
sub.meta=full.meta
sub.meta[sub.meta$species %in% c("O.robusta","M.geometrizans"),"species"]="Cacti" #merge cacti samples
result_norm=as.data.frame(t(t(data.full)/colSums(data.full))) #relative abundance
result_norm$ko_id=rownames(result_norm) #ko names
result.tidy=gather(result_norm, key = sample, value = rel.abundance, keys[files]) #create tidy data
for (i in keys){
result.tidy[result.tidy$sample==i,c(4,5,6)]=sub.meta[sub.meta$Genome_name==i,c("species","site","compartment")]
} #add metadata
# SELECT DATA
file=c("Expanded_families_edgeR.txt","Reduced_families_edgeR.txt","Comparative_Genomics_families_edgeR.txt")
comparison=c("RHI_Prokaryotes_ko_0.75", "PHY_Prokaryotes_ko_0.75")[h] #select the comparison
if (sum(grep("RHI",comparison))>0){
pair=c("salmiana_vs_tequilana","salmiana_vs_cacti", "cacti_vs_tequilana")
datos=result.tidy[result.tidy$compartment == "rhizosphere",]
} else if (sum(grep("PHY",comparison)>0)){
pair=c("salmiana_vs_tequilana","salmiana_vs_cacti", "cacti_vs_tequilana")
datos=result.tidy[result.tidy$compartment == "phyllosphere",]
} #select tha data from the tidy table based on the comparison
sub1.1=read.delim(paste(main.dir,comparison,pair[1],file[1], sep = "/"))
sub1.2=read.delim(paste(main.dir,comparison,pair[1],file[2], sep = "/"))
sub2.1=read.delim(paste(main.dir,comparison,pair[2],file[1], sep = "/"))
sub2.2=read.delim(paste(main.dir,comparison,pair[2],file[2], sep = "/"))
sub3.1=read.delim(paste(main.dir,comparison,pair[3],file[1], sep = "/"))
sub3.2=read.delim(paste(main.dir,comparison,pair[3],file[2], sep = "/")) #read the expanded[1] and reduced[2] KOs per comparison [1,2,3]
# TRANSFORM DATA
mean.data=summarySE(data=datos, measurevar = "rel.abundance", groupvars = c("ko_id","species")) #calculate mean relative
mean.data2=mean.data[,c(1,2,4)] #select necessary data
plot.data=spread(mean.data2, key = species, value = rel.abundance) #transform data from tidy to untidy in order to have the 3 axis for the triplot
# SUBSET ONLY COMPARED KOs
all=read.delim(paste(main.dir,comparison,pair[1],file[3], sep = "/")) #read all the compared KOs
plot.data=plot.data[plot.data$ko_id %in% rownames(all),] #select them from data
# LABEL SIGNIFICANT DATA
plot.data$value="No significant"
plot.data[plot.data$ko_id %in% c(rownames(sub1.1),rownames(sub1.2),rownames(sub2.1),rownames(sub2.2),rownames(sub3.1),rownames(sub3.2)),"value"]="Significant"
# HIGHLIGHT PROCESSES
sub.process=process[names(process)[c(1:6)]]
sub.aes=aes3[aes3$ruta%in%names(sub.process),]
plot.data$ruta="Other"
for (i in rev(names(sub.process))){
plot.data[plot.data$ko_id %in% sub.process[[i]], "ruta"]=i
} # label each KO with a process
plot.data$ruta=factor(plot.data$ruta, levels = c("Other",names(sub.process)))
# ADD FDR
all=all[plot.data$ko_id,]
plot.data$p.value=all$FDR #non significant FDR will be based on the first comparison (pair[1])
p=rbind(sub1.1,sub1.2,sub2.1,sub2.2,sub3.1,sub3.2)
for (i in rownames(p)){
plot.data[plot.data$ko_id==i,"p.value"]=p[i,"FDR"]
} #subtitute the FDR value of significan data of each comparison
# CREATE CHEAT DATSET TO HIGHLIGHT SPECIFIC POINTS
cheat=plot.data[plot.data$ruta %in% names(sub.process) & plot.data$value == "Significant",]
cheat2=plot.data[plot.data$ruta %in% names(sub.process)[6] & plot.data$value == "Significant",]
# PLOT
a=ggtern(data=plot.data,aes(x=A.salmiana, y=A.tequilana, z=Cacti, shape=value, color=ruta, size=-log10(p.value)))+
geom_mask()+
geom_point()+
geom_point(data =cheat, aes(x=A.salmiana, y=A.tequilana, z=Cacti, shape=value, color=ruta, size=-log10(p.value)))+
geom_point(data =cheat2, aes(x=A.salmiana, y=A.tequilana, z=Cacti, shape=value, color=ruta, size=-log10(p.value)))+
scale_shape_manual(values=c(21,19))+
scale_size_continuous(range = c(1,2))+
scale_color_manual(values =c("grey70",as.character(aes3[aes3$ruta %in% names(sub.process),1])))+
ylab("")+xlab("")+zlab("")+tema
# SAVE
png(filename = paste(result.dir,paste(comparison,"ternary_9","png",sep = "."),sep = "/"),
width = 160, height = 100, units = "mm", pointsize = 15, bg = "white", res=300, type="cairo")
print(a) ; dev.off()
}