-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathmake_covars_cv.R
173 lines (137 loc) · 8.52 KB
/
make_covars_cv.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
.libPaths("/moto/palab/users/zlf2101/rpackages/")
library(SNPRelate)
require(sp)
##READ INDEX NUMBER FOR JACKKNIFE TRIAL
args <- commandArgs(TRUE)
trial_index <- args[1]
##MAKE COVARIATES FOR BLEACHING GWAS - CV JACKNIFE
##READ IN SAMPLE DATA
#full_pheno<-read.table("~/Amil.chl.missing_rmvd.plink.pheno")
full_pheno<-read.table("amil_nomissing.zoox.pheno")
names(full_pheno)<-c("IID","FID","chl_content")
full_pheno$bleached<-ifelse(full_pheno$chl_content>3,2,1)
full_pheno$Reef<-substr(full_pheno$IID,start=1,stop=2)
##SPLIT DATA INTO TRAIN/TEST
smp_siz = floor(0.90 * nrow(full_pheno))
train_ind = sample(seq_len(nrow(full_pheno)),size=smp_siz)
train = full_pheno[train_ind,]
test = full_pheno[-train_ind,]
train = full_pheno
##MAKE ENVIRONMENTAL PCS FOR COVARIATES
train_reefs<-unique(train$Reef)
test_reefs<-unique(test$Reef)
coords<-read.csv("data/reef_gps_coords.csv",colClasses="character")
coords$Lat<-sub('\xa1', 'd', coords$Lat)
coords$Lat<-as.numeric(sp::char2dms(coords$Lat))
coords$Lon<-sub('\xa1', 'd', coords$Lon)
coords$Lon<-as.numeric(sp::char2dms(coords$Lon))
coords<-coords[order(coords$Reef),]
names(coords)<-c("Reef","Latitude","Longitude")
env_dat<-read.csv("data/Environmental_data.csv")
dhw_dat<-read.csv("data/DHW_data.csv")
datalist = list()
for(i in 1:nrow(coords)){
dists<-( (coords[i,]$Longitude-env_dat$lon)*cos(90*env_dat$lat ) )^2 + (coords[i,]$Latitude-env_dat$lat)^2
dhw_dists<-( (coords[i,]$Longitude-dhw_dat$lon)*cos(90*dhw_dat$lat ) )^2 + (coords[i,]$Latitude-dhw_dat$lat)^2
env_row<-env_dat[ which.min(dists),]
dhw_row<-dhw_dat[ which.min(dhw_dists),]$annMaxDHW_2017
env_row$DHW_max<-dhw_row
datalist[[i]] <- env_row
}
reef_dat = do.call(rbind, datalist)
reef_dat<-cbind(coords,reef_dat)
reef_dat[,c("PIXEL_ID","lon","lat","REEF_ID","REEF_NAME")]<-NULL
head(reef_dat)
drop_cols<-c("pop","FAMID","SECTOR","CROSS_SHELF","Comp.1","Comp.2","Comp.3","Comp.4","Comp.5","DHW_max","Latitude","Longitude","gcta_fid")
env_covars_names<-c("CRS_NO3_AV","CRS_NO3_SR","CRS_PO4_AV","CRS_PO4_SR","CRS_O2_AV","CRS_O2_SR","CRS_S_AV","CRS_S_SR","CRS_T_AV","CRS_T_SR","CRS_SI_AV","CRS_SI_SR",
"GA_BATHY","GA_SLOPE","GA_ASPECT","GBR_BATHY","GA_CRBNT","GA_GRAVEL","GA_SAND","GA_MUD","GMCS_STRESS_TMN","GMCS_STRESS_IQR",
"SW_CHLA_AV","SW_CHLA_SR","SW_K490_AV","SW_K490_SR","SW_BIR_AV","SW_BIR_SR","MT_SST_AV","MT_SST_SR","MT_SST_MIN","Primary","Secondary","Tertiary","Latitude","Longitude",
"DHW_max","mindistbar","mindistcoa")
rownames(reef_dat)<-reef_dat$Reef
env_pca<-prcomp(reef_dat[,env_covars_names], center = TRUE,scale. = TRUE)
env_pca_loadings<-data.frame(env_pca$x)
env_pca_loadings$Reef<-as.vector(as.character(rownames(env_pca_loadings)))
##READ IN DATA FOR MISSING READ PROPORTION
missing_p<-read.table("data/missing.smiss")
head(missing_p)
missing_p<-missing_p[,c("V1","V4")]
names(missing_p)<-c("IID","prop_missing")
train_covars<-mermage(train, missing_p, by="IID")perm
test_covars<-merge(test, missing_p, by="IID")
train_covars<-merge(train_covars,env_pca_loadings,by="Reef",all.x=T)
test_covars<-merge(test_covars,env_pca_loadings,by="Reef",all.x=T)
locov_read_counts<-read.table("data/locov_merged_read_counts.tsv",header=T)
hicov_read_counts<-read.table("data/merged_read_counts.tsv",header=T)
scaff_lengths<-read.table("data/scaffold_names_lenghts.tsv")
names(scaff_lengths)<-c("scaff","length")
amil_scaffs<-scaff_lengths$scaff[!(scaff_lengths$scaff %in% c("symbiont1","symbiont2","symbiont3","symbiont4","Acropora_millepora_mitochondria"))]
locov_reads<-(locov_read_counts[locov_read_counts$scaff %in% amil_scaffs,2:length(locov_read_counts)])
hicov_reads<-(hicov_read_counts[hicov_read_counts$scaff %in% amil_scaffs,2:length(hicov_read_counts)])
locov_total_read_sums<-colSums(locov_reads)
hicov_total_read_sums<-colSums(hicov_reads)
total_read_counts<-c(hicov_total_read_sums,locov_total_read_sums)
total_read_counts<-data.frame(as.table(total_read_counts))
head(total_read_counts)
names(total_read_counts)<-c("ID","total_read_count")
train_covars<-merge(train_covars,total_read_counts,by.y="ID",by.x="IID")
test_covars<-merge(test_covars,total_read_counts,by.y="ID",by.x="IID")
head(test_covars)
quantNorm =function(x){qnorm(rank(x,ties.method = "average")/(length(x)+1))}
train_covars$norm_read_count<-quantNorm(train_covars$total_read_count)
all_read_counts<-rbind(train_covars[,c("IID","total_read_count")],test_covars[,c("IID","total_read_count")])
all_read_counts$norm_read_counts<-quantNorm(all_read_counts$total_read_count)
test_norm_reads<-all_read_counts[all_read_counts$IID %in% test_covars$IID,]
test_covars$norm_read_count<-test_norm_reads$norm_read_counts
##GENETIC PCS
genofile<-snpgdsOpen("data/Amil.full_samps.vcf.gds", readonly=FALSE, allow.duplicate = T)
train_snpset <- snpgdsLDpruning(genofile, ld.threshold=0.2,sample.id = as.factor(train_covars$IID))
train_snpset.id <- unlist(train_snpset)
train_pca <- snpgdsPCA(genofile, snp.id=train_snpset.id, num.thread=2,sample.id = as.factor(train_covars$IID),need.genmat=TRUE)
train_pca_res<-prcomp(train_pca$genmat)
train_tab <- data.frame(sample.id = train_pca$sample.id,
EV1 = train_pca$eigenvect[,1], # the first eigenvector
EV2 = train_pca$eigenvect[,2],
EV3 = train_pca$eigenvect[,3],
EV4 = train_pca$eigenvect[,4],# the second eigenvector
stringsAsFactors = FALSE)
head(train_tab)
train_covars<-merge(train_covars,train_tab,by.x="IID",by.y="sample.id")
train_snp_loadings<-snpgdsPCASNPLoading(train_pca, genofile)
test_loadings<-snpgdsPCASampLoading(train_snp_loadings,genofile,sample.id=as.factor(test_covars$IID))
test_tab <- data.frame(sample.id = test_loadings$sample.id,
EV1 = test_loadings$eigenvect[,1], # the first eigenvector
EV2 = test_loadings$eigenvect[,2],
EV3 = test_loadings$eigenvect[,3],
EV4 = test_loadings$eigenvect[,4],# the second eigenvector
stringsAsFactors = FALSE)
test_covars<-merge(test_covars,test_tab,by.x="IID",by.y="sample.id")
##ADD DATES
sample_dates<-read.table("data/Amil_sample_dates.txt",sep="\t",header=T)
train_covars<-merge(sample_dates[c("SampleID","Date","Field.ID")],train_covars,by.x = "SampleID",by.y="IID")
train_covars$Date<-as.character(train_covars$Date)
train_covars$Date<-as.numeric(as.Date(train_covars$Date, format = "%m/%d/%y",origin="1970-01-01 00:00:00 +10:00"))
min_date<-min(train_covars$Date)
train_covars$Date<-train_covars$Date - min_date
test_covars<-merge(sample_dates[c("SampleID","Date","Field.ID")],test_covars,by.x = "SampleID",by.y="IID")
test_covars$Date<-as.character(test_covars$Date)
test_covars$Date<-as.numeric(as.Date(test_covars$Date, format = "%m/%d/%y",origin="1970-01-01 00:00:00 +10:00"))
test_covars$Date<-test_covars$Date - min_date
##ADD DEPTH
depth_data<-read.table("data/Amil_depth.txt",sep="\t",header=T)
colnames(train_covars)[colnames(train_covars) == "SampleID"] <- "IID"
train_covars<-merge(depth_data,train_covars,by="IID")
colnames(test_covars)[colnames(test_covars) == "SampleID"] <- "IID"
test_covars<-merge(depth_data,test_covars,by="IID")
##ADD SYMBIONT DATA
chl_reg<-read.table("data/predicted2_chl_vals.tsv",sep="\t",header=T)
colnames(chl_reg)[colnames(chl_reg) == "SampleID"] <- "IID"
train_covars<-merge(train_covars,chl_reg,by="IID")
test_covars<-merge(test_covars,chl_reg,by="IID")
##WRITE FILES
write.table(train_covars[,c("FID","IID","Dom_symbiont","PC1","PC2","PC3","PC4","PC5","PC6","PC7","PC8","EV1","EV2","norm_read_count","prop_missing","Depth","Date")],paste("cv_input/train_covars_",trial_index,".txt",sep=""),row.names=F,quote=F)
write.table(test_covars[,c("FID","IID","Dom_symbiont","PC1","PC2","PC3","PC4","PC5","PC6","PC7","PC8","EV1","EV2","norm_read_count","prop_missing","Depth","Date")],paste("cv_input/test_covars_",trial_index,".txt",sep=""),row.names=F,quote=F)
write.table(train_covars[,c("FID","IID","bleached")],paste("cv_input/train_pheno_",trial_index,".txt",sep=""),quote=F,row.names = F,col.names = F)
write.table(train_covars[,c("FID","IID")],paste("cv_input/plink19_gwas_set_",trial_index,".txt",sep=""),quote=F,row.names = F,col.names = F)
write.table(test_covars[,c("FID","IID")],paste("cv_input/plink19_validation_set_",trial_index,".txt",sep=""),quote=F,row.names = F,col.names = F)
write.table(train_covars[,c("IID")],paste("cv_input/plink2_gwas_set_",trial_index,".txt",sep=""),quote=F,row.names = F,col.names = F)
write.table(test_covars[,c("IID")],paste("cv_input/plink2_validation_set_",trial_index,".txt",sep=""),quote=F,row.names = F,col.names = F)