install.packages ("tydiverse", independencies = T)

library(vegan)
library(tidyverse)


####Fungi(ITS)#######

Read in the relevant files:
```{r, cache = TRUE}
OTUS_ITS <- read.csv("input/OTU_table.csv")
head (OTUS_ITS)

tax_ITS <- read.csv("input/ITS_taxonomy_clean.csv")
head (tax_ITS)


##################Rarefaction########################

plot(log(sort(colSums(OTUS_ITS[,-1]), decreasing = T), base = 10))

OTUS_MITS <- as.matrix(OTUS_ITS[,-1])
dimnames(OTUS_MITS) <- list(OTUS_ITS[,1], colnames(OTUS_ITS[,-1]))

head(OTUS_MITS)


rar_Richness_ITS <- rrarefy(OTUS_MITS, min(colSums(OTUS_MITS)))
head(rar_Richness_ITS)
tail(rar_Richness_ITS)


write.csv(rar_Richness_ITS, "otu_ITS_rarrefied_soil.csv")

srar_Richness_ITS <- rarefy(t(OTUS_MITS), sample = min(colSums(OTUS_MITS)), se = T)
head(srar_Richness_ITS)

Richness_ITS <- as.matrix(srar_Richness_ITS)
head(Richness_ITS)


write.csv(Richness_ITS, "rich_ITS_rarrefied.csv")

############Figure S1A

pdf('FigS1A_ITS_raref.pdf')
rarecurve(t(OTUS_MITS), step = 1000, label = "Fungi - ITS", ylab = "OTU", title =  "Fungi - ITS")
dev.off()


#########################################################################
####################Taxonomic Assignment#################################
#########################################################################

############################################################
#ITS
############################################################
##load and prepare data
###taxonomy
tax <- read_csv("input/ITS_taxonomy_clean.csv")
tax <- tax%>%
  dplyr::select(OTU.ID = OTU_ID, taxonomy=classifications)

View(tax)
head(tax)
tail(tax)


###abundance per site

#soil
abus <- read_csv("otu_ITS_rarrefied_soil.csv")
head(abus)

abus <- abus%>%
rename(OTU.ID = X1)

nams <- abus$OTU.ID

abus[abus > 1] <-1 #convert to presence/absence
abus$OTU.ID <- nams


##merge taxonomy and abundance
abutaxs <- full_join(abus, tax, by = "OTU.ID")
head (abutaxs)

write.csv(abutaxs, "input/ITS_Fungi_taxonomy_metadata2.csv")

###Identity and number of OTUs not in the taxonomy file
sum(is.na(abutaxs$taxonomy))
sum(is.na(abutaxs$taxonomy))/nrow(abus)


###Identity and number of OTUs not in the taxonomy file
sum(is.na(abutaxs$SBCIGP1))
sum(is.na(abutaxs$SBCIGP1))/nrow(tax)


##Create taxonomy table 
###not rarefied
####put table in right format for summary
snon_rar_ITS <- gather(abutaxs, key = site, value = presence, -OTU.ID, -taxonomy)%>%
  filter(presence > 0)


####define the taxonomic level to use for all groups,and add to table
tax.list <- c("p__Ascomycota","p__Basidiomycota","p__Chytridiomycota","p__Entorrhizomycota","p__Glomeromycota","p__Kickxellomycota",
              "p__Monoblepharomycota","p__Mortierellomycota","p__Mucoromycota","p__Mucoromycotina","p__Neocallimastigomycota",
              "p__Rozellomycota","p__Zoopagomycota","p__unidentified")


snon_rar_ITS <- snon_rar_ITS%>%
  mutate(tax.class = "others")%>%
  mutate(tax.class = ifelse(grepl(tax.list[1], taxonomy), tax.list[1], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[2], taxonomy), tax.list[2], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[3], taxonomy), tax.list[3], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[4], taxonomy), tax.list[4], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[5], taxonomy), tax.list[5], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[6], taxonomy), tax.list[6], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[7], taxonomy), tax.list[7], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[8], taxonomy), tax.list[8], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[9], taxonomy), tax.list[9], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[10], taxonomy), tax.list[10], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[11], taxonomy), tax.list[11], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[12], taxonomy), tax.list[12], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[13], taxonomy), tax.list[13], tax.class))%>%
  mutate(tax.class = ifelse(grepl(tax.list[14], taxonomy), tax.list[14], tax.class))%>%
mutate(tax.class = ifelse(is.na(taxonomy), "unknown", tax.class))


####create output tables
head(abus)
#create site to locality and habitat key
skeys <- names(abus)%>%
  as_tibble()%>%
  rename(site = value)%>%
  mutate(locality = "BC")%>%
  mutate(locality = ifelse(grepl("CUI", site), "CUI", locality))%>%
  mutate(locality = ifelse(grepl("CXN", site), "CXN", locality))%>%
  mutate(locality = ifelse(grepl("JAU", site), "JAU", locality))%>%
    mutate(habitat = "IG")%>%
  mutate(habitat = ifelse(grepl("TF", site), "TF", habitat))%>%
  mutate(habitat = ifelse(grepl("VZ", site), "VZ", habitat))%>%
  mutate(habitat = ifelse(grepl("CAM", site), "CAM", habitat))%>%
    mutate(sample = "In")%>%
  mutate(sample = ifelse(grepl("L", site), "L", sample))%>%
  mutate(sample = ifelse(grepl("S", site), "S", sample))


out_abus <- snon_rar_ITS%>%
  left_join(skeys, by = "site")



tail(out_abus)


write_csv(out_abus, "ITS_taxonomy_per_site_rarefied.csv")


######################################################################
#####################Plots taxonomic groups############################
######################################################################

##########################################
#rarified
##########################################

ITS_rars <- read_csv("ITS_taxonomy_per_site_rarefied.csv")

head(ITS_rars)
ITS_rars$locality


#load ITS
plot_ITS <- ITS_rars%>%
  distinct(OTU.ID, site, .keep_all = T)%>%
  group_by(site, tax.class)%>%
  summarise(OTUs.number = n()) %>%
  mutate(fraction.OTUs = round(OTUs.number / sum(OTUs.number), 3))%>%
  filter(!is.na(site))%>%
  rename(value = site)%>%
  mutate(type = "site")
write.csv(plot_ITS, "plot_ITS.csv")
plot_ITS <- read.csv("plot_ITS.csv")

samp_ITS <- ITS_rars%>%
  distinct(OTU.ID, sample, .keep_all = T)%>%
  group_by(sample, tax.class)%>%
  summarise(OTUs.number = n()) %>%
  mutate(fraction.OTUs = round(OTUs.number / sum(OTUs.number), 3))%>%
  filter(!is.na(sample))%>%
  rename(value = sample)%>%
  mutate(type = "sample")

sloc_ITS <- ITS_rars%>%
  distinct(OTU.ID, locality, .keep_all = T)%>%
  group_by(locality, tax.class)%>%
  summarise(OTUs.number = n()) %>%
  mutate(fraction.OTUs = round(OTUs.number / sum(OTUs.number), 3))%>%
  filter(!is.na(locality))%>%
  rename(value = locality)%>%
  mutate(type = "locality")


splo_ITS <- bind_rows(plot_ITS,samp_ITS)%>%
  mutate(marker = "ITS")
#View(splo_ITS)
write.csv(splo_ITS, "figure.csv")

splo_ITS$tax.class <-factor(splo_ITS$tax.class, levels = unique(splo_ITS$tax.class[order(splo_ITS$fraction.OTUs)]))


#number 

mycolors = c(brewer.pal(name="Dark2", n = 8), brewer.pal(name="Paired", n = 6))

ITS_plot <- ggplot(data = plot_ITS, aes(x= value, y=OTUs.number, fill=tax.class))+
  geom_bar(stat = "identity", position="dodge")+
  scale_color_manual(values = "mycolors", name = "Taxa Fungi (ITS) Soil")+
  ylab("Number of OTUs")+
  xlab("")+
  coord_flip()+
  theme_bw()+
  theme(panel.grid.major = element_blank(), 
panel.grid.minor = element_blank(),
panel.background = element_blank(),
axis.line = element_line(colour = "black"),
panel.border = element_rect(colour = "black", fill=NA, size=1),
        legend.position = "right")+
  facet_grid(~type, scales = "free_x")



ggsave("ITS_number_of_OTUs_perPlot4.pdf", plot =ITS_plot, width =8, height =12)

#fraction

ITS_S <- ggplot(data = splo_ITS, aes(x= value, y=fraction.OTUs, fill=tax.class))+
  geom_bar(stat = "identity", position=position_dodge())+
  scale_color_manual(values = "mycolors", name = "Taxa Prokaryote (ITS) Soil")+
  ylab("Fraction of OTUs")+
  xlab("")+
  theme_bw()+
  theme(panel.grid.major = element_blank(), 
panel.grid.minor = element_blank(),
panel.background = element_blank(),
axis.line = element_line(colour = "black"),
panel.border = element_rect(colour = "black", fill=NA, size=1),
        legend.position = "right")+
  facet_grid(~type, scales = "free_x")


ggsave("ITS_fraction_of_OTUs_rarefied.pdf", plot = ITS_S,  width =12, height = 8)



#########################################
#OTU "ranges"
#################################################

ITS_unrar <- read_csv("ITS_taxonomy_per_site_rarefied.csv")


#ITS
pn <- ITS_unrar%>%
  select(-taxonomy, -locality, -habitat)%>%
  group_by(tax.class, OTU.ID)%>%
  summarise(plot_number = n())%>%
  filter(!tax.class == "unknown")%>%
  filter(!tax.class == "others")%>%
  ungroup()%>%
  mutate(tax.class = gsub("bacteria", "bac", tax.class))

#number of habitats
hab_n <- 
  ITS_unrar%>%
  filter(!tax.class == "unknown")%>%
  filter(!tax.class == "others")%>%
  filter(!is.na(habitat))%>%
  select(-taxonomy, -locality)%>%
  group_by(tax.class, OTU.ID, habitat)%>%
  summarise(OTUs.number = n())%>%
  ungroup()%>%
  select(-OTUs.number)%>%
  group_by(tax.class, OTU.ID)%>%
  summarise(habitat_number = n())%>%
  group_by(tax.class, habitat_number )%>%
  summarise(num_hab_num = n())%>%
  mutate(fraction_hab_num = round(num_hab_num/ sum(num_hab_num), 3))%>%
  ungroup()%>%
  mutate(tax.class = gsub("bacteria", "bac", tax.class))


#number of localities
loc_n <- 
  ITS_unrar%>%
  filter(!tax.class == "unknown")%>%
  filter(!tax.class == "others")%>%
  filter(!is.na(locality))%>%
  select(-taxonomy, -habitat)%>%
  group_by(tax.class, OTU.ID, locality)%>%
  summarise(OTUs.number = n())%>%
  ungroup()%>%
  select(-OTUs.number)%>%
  group_by(tax.class, OTU.ID)%>%
  summarise(locality_number = n())%>%
  group_by(tax.class, locality_number )%>%
  summarise(num_loc_num = n())%>%
  mutate(fraction_loc_num = round(num_loc_num/ sum(num_loc_num), 3))%>%
  ungroup()%>%
  mutate(tax.class = gsub("bacteria", "bac", tax.class))


plo_pn <- ggplot()+
  geom_bar(aes(x = taxonomy, y = fraction, fill = sample),
           stat = "identity", position = "dodge")+
  ylab("Number of sites")+
  ggtitle("Number of sites")+
  theme_bw()+
  theme(panel.grid.major = element_blank(), 
panel.grid.minor = element_blank(),
panel.background = element_blank(),
axis.line = element_line(colour = "black"),
panel.border = element_rect(colour = "black", fill=NA, size=1))

plo_pn_den <-ggplot()+
  geom_density(data  = pn, aes(plot_number, group = tax.class, col = tax.class))+
  scale_colour_brewer(palette = "Paired", name = "")+
  ggtitle("Number of sites - density")+
  xlab("Number of sites")+
  ylab("Density")+
  xlim(c(1, 39))+
  guides(color=guide_legend(ncol=2))+
    theme_bw()+
  theme(panel.grid.major = element_blank(), 
panel.grid.minor = element_blank(),
panel.background = element_blank(),
axis.line = element_line(colour = "black"),
panel.border = element_rect(colour = "black", fill=NA, size=1),
legend.position = c(0.65,0.78),
        legend.title = element_blank())


plo_habn <- ggplot(data = hab_n, aes(x= tax.class, y=fraction_hab_num, fill=as.factor(habitat_number)))+
  geom_bar(stat = "identity", position=position_dodge())+
  scale_fill_brewer(palette = "Set3", name = "")+
  ylab("Fraction of OTUs")+
  ggtitle("Number of habitats")+
  guides(fill=guide_legend(ncol=4))+
  theme_bw()+
  theme(panel.grid.major = element_blank(), 
panel.grid.minor = element_blank(),
panel.background = element_blank(),
axis.line = element_line(colour = "black"),
panel.border = element_rect(colour = "black", fill=NA, size=1),
        legend.position = c(0.8, 0.93),
        legend.title = element_blank())

plo_locn <- ggplot(data = loc_n, aes(x= tax.class, y=fraction_loc_num, fill=as.factor(locality_number)))+
  geom_bar(stat = "identity", position=position_dodge())+
  scale_fill_brewer(palette = "Accent", name = "")+
  ggtitle("Number of localities")+
  ylab("Fraction of OTUs")+
  guides(fill=guide_legend(ncol=4))+
  theme_bw()+
  theme(panel.grid.major = element_blank(), 
panel.grid.minor = element_blank(),
panel.background = element_blank(),
axis.line = element_line(colour = "black"),
panel.border = element_rect(colour = "black", fill=NA, size=1),
        legend.position = c(0.8, 0.93),
        legend.title = element_blank())

out <- grid.arrange(plo_pn, plo_pn_den, plo_habn, plo_locn, top = "OTU range size ITS")
library(gridExtra)

ggsave("Otu_ranges_ITS_rarified.pdf", plot = out, width = 12, height = 8)

#########################################################################
################Making the presence/absence matrix#######################
#########################################################################

comITS  <- rar_Richness_ITS 
comITS [comITS  > 1] <- 1
comITS <- as.data.frame(comITS)
nrow (comITS)
write.csv(comITS , "otu_ITS_presence_absence.csv")

##########Betadiversity analysis

comITS <- comITS%>%
  filter(rowSums(comITS) != 0)

ITS.nmds <- metaMDS(as.matrix(t(comITS)), distance="jaccard",k=2, autotransform=F)
ITS.nmds <- metaMDS(as.matrix(t(comITS)), distance="jaccard",k=2, autotransform=F, previou.best = ITS.nmds)
summary(ITS.nmds)
scITS <- data.frame(scores(ITS.nmds))
scITS$ID <- rownames(scITS)
write.csv(scITS, "ITS_NMDS_metadata.csv")

biol <- read.csv("input/metadata.csv")

scITS <- scITS %>%
  as_data_frame()%>%
  inner_join(biol, by = "ID")

##envfit

###Eukaryote(ITS)
envfit(scITS[,1:2] ~ as.factor(scITS$Locality))
envfit(scITS[,1:2] ~ as.factor(scITS$Habitat))
envfit(scITS[,1:2] ~ as.factor(scITS$Sample))


cols1 <- c("#e66101",
           "#fdb863",
           "#b2abd2",
           "#5e3c99")

cols2 <- c("#ca0020",
           "#dfc27d",
           "#92c5de",
           "#0571b0")


figA <- scITS %>%
  ggplot()+
  geom_point(aes(x = NMDS1, y = NMDS2, colour = Habitat, shape = Sample), size = 3)+
  ggtitle("A - ITS fo Habitat")+
  xlab("")+
  ylab("")+
  scale_fill_discrete(name = "Environmental\ntype")+
  scale_color_manual(values=cols1)+
  theme(panel.grid.major = element_blank(), 
        panel.grid.minor = element_blank(),
        panel.background = element_blank(),
        axis.line = element_line(colour = "black"),
        panel.border = element_rect(colour = "black", fill=NA, size=1),
        legend.position = ("left"))


#subfigure B
figB <- scITS %>%
  ggplot()+
  geom_point(aes(x = NMDS1, y = NMDS2, colour = Locality, shape = Sample), size = 3)+
  ggtitle("B - ITS fo Locality")+
  xlab("")+
  ylab("")+
  scale_fill_discrete(name = "Environmental\ntype")+
  scale_shape_manual(values=c(16:17))+
  scale_color_manual(values=cols2)+
  theme(panel.grid.major = element_blank(), 
        panel.grid.minor = element_blank(),
        panel.background = element_blank(),
        axis.line = element_line(colour = "black"),
        panel.border = element_rect(colour = "black", fill=NA, size=1),
        legend.position = ("left"))



fig2 <- gridExtra::grid.arrange(figA,figB, bottom = "NMDS1", left = "NMDS2")
ggsave("Fig_NMDS_Habitat_and_Locality.pdf", plot = fig2, width =10, height = 12)

disITS  <- vegdist(as.matrix(t(comITS)), method = "jaccard")

summary(disITS)

#########################################################
###################PERMANOVA#############################
#########################################################

tcomITS <- t(comITS)
tcomITS <- as.data.frame (tcomITS)


adoITS <- adonis2(tcomITS ~ biol$Locality+ biol$Habitat + biol$Sample, strata = biol$ID, data = tcomITS, type = 'Jaccard', binary = TRUE)
summary(adoITS)
adoITS 

adoITS2 <- adonis(tcomITS ~ biol$Locality+ biol$Habitat + biol$Sample, strata = biol$ID, data = tcomITS, type = 'Jaccard', binary = TRUE)
adoITS2

################################################################
##################Variance partitioning#########################
################################################################

mod <- varpart(disITS, ~ biol$Locality, ~ biol$Sample, ~ biol$Habitat)
mod
plot(mod)

rda.all <- rda(t(comITS) ~ biol$Locality + biol$Sample, data = biol)
anova(rda.all)

rda.sample <- rda(t(comITS) ~ Sample, data = biol)
anova(rda.sample)

rda.locality <- rda(t(comITS) ~ Locality, data = biol)
anova(rda.locality)

rda.habitat <- rda(t(comITS) ~ Habitat, data = biol)
anova(rda.habitat)