/
7_ProcessingResults.R
272 lines (238 loc) · 11.6 KB
/
7_ProcessingResults.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
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
## Companion script for the manuscript:
## "Estimation of Molecular Pairwise Relatedness in Autopolyploid Crops"
## by Rodrigo Amadeu, Leticia Lara, Patricio Munoz, Antonio Garcia
## Sep-2020
## This script:
## computes the metrics between observed and estimated relationships and save them in a table format.
library(DescTools) #for CCC coefficient
reps = 3
## Creating a data.frame to organize the results
stat.PolyRelatedness <- expand.grid(
methods = c("LO","RI","WE","MM","ML"),
ploidy = c("2ploidy","4ploidy","6ploidy","8ploidy","4ploidy_natural","6ploidy_natural","8ploidy_natural"),
loci = c("5","10","15","20","50","100","500","1000"),
alleles = c("2alleles","3alleles","4alleles","5alleles","10alleles","15alleles","20alleles"),
RMSE.PO = NA,
RMSE.FS = NA,
RMSE.HS = NA,
RMSE.GPGO = NA,
RMSE.UN = NA,
RMSE.GUGN = NA,
RMSE.UR = NA,
IC.PO = NA,
IC.FS = NA,
IC.HS = NA,
IC.GPGO = NA,
IC.UN = NA,
IC.GUGN = NA,
IC.UR = NA,
meancor.Farthing = NA,
meancor.IAC = NA,
RMSE.Farthing = NA,
RMSE.IAC = NA,
CCC.Farthing = NA,
CCC.IAC = NA,
CCC.lwrci.Farthing = NA,
CCC.uprci.Farthing = NA,
CCC.lwrci.IAC = NA,
CCC.uprci.IAC = NA)
stat.AGHmatrix <- expand.grid(
methods = c("VanRaden","Yang","Slater"),
ploidy = c("2ploidy","4ploidy","6ploidy","8ploidy","4ploidy_natural","6ploidy_natural","8ploidy_natural"),
loci = c("5","10","15","20","50","100","500","1000"),
alleles = c("2alleles"),
RMSE.PO = NA,
RMSE.FS = NA,
RMSE.HS = NA,
RMSE.GPGO = NA,
RMSE.UN = NA,
RMSE.GUGN = NA,
RMSE.UR = NA,
IC.PO = NA,
IC.FS = NA,
IC.HS = NA,
IC.GPGO = NA,
IC.UN = NA,
IC.GUGN = NA,
IC.UR = NA,
meancor.Farthing = NA,
meancor.IAC = NA,
RMSE.Farthing = NA,
RMSE.IAC = NA,
CCC.Farthing = NA,
CCC.IAC = NA,
CCC.lwrci.Farthing = NA,
CCC.uprci.Farthing = NA,
CCC.lwrci.IAC = NA,
CCC.uprci.IAC = NA)
## Extracting observed relatedness
obs.r = list()
load("ObservedRelatedness/r2ploidy.Rdata")
obs.r$'2ploidy' = mat.array
load("ObservedRelatedness/r4ploidy.Rdata")
obs.r$'4ploidy' = mat.array
load("ObservedRelatedness/r6ploidy.Rdata")
obs.r$'6ploidy' = mat.array
load("ObservedRelatedness/r8ploidy.Rdata")
obs.r$'8ploidy' = mat.array
load("ObservedRelatedness/rnatural4ploidy.Rdata")
obs.r$'4ploidy_natural' = mat.array
load("ObservedRelatedness/rnatural6ploidy.Rdata")
obs.r$'6ploidy_natural' = mat.array
load("ObservedRelatedness/rnatural8ploidy.Rdata")
obs.r$'8ploidy_natural' = mat.array
## Extracting results and computing metrics for PolyRelatedness metrics
## "LO","RI","WE","MM","ML"
for(i in 1:nrow(stat.PolyRelatedness)){
print(paste0(i,"/",nrow(stat.PolyRelatedness)))
est.r = obs.r$'2ploidy'*0 #creates an empity array
## loop to extract the PolyRelatedness results
for(j in 1:reps){
mat = read.table(paste0("PolyRelatedness_output_",stat.PolyRelatedness$methods[i],"/",
stat.PolyRelatedness$ploidy[i],"/",
stat.PolyRelatedness$loci[i],"loci/",
stat.PolyRelatedness$alleles[i],"/",
j,".txt"),skip=7)
mat = as.matrix(mat)
est.r[,,j] = mat
}
ploidy = as.character(stat.PolyRelatedness$ploidy[i])
## gather observed relatedness (r)
PO.obs <- obs.r[[ploidy]][181,182,1:reps]
FS.obs <- obs.r[[ploidy]][179,180,1:reps]
HS.obs <- obs.r[[ploidy]][180,181,1:reps]
GPGO.obs <- obs.r[[ploidy]][181,183,1:reps]
UN.obs <- obs.r[[ploidy]][180,182,1:reps]
GUGN.obs <- obs.r[[ploidy]][180,183,1:reps]
Farthing.obs <- obs.r[[ploidy]][74:149,150,1:reps]
IAC.obs <- obs.r[[ploidy]][151:176,177,1:reps]
## gather estimated relatedness (r)
PO.est <- est.r[181,182,1:reps]
FS.est <- est.r[179,180,1:reps]
HS.est <- est.r[180,181,1:reps]
GPGO.est <- est.r[181,183,1:reps]
UN.est <- est.r[180,182,1:reps]
UR.est <- est.r[1:10,1:10,1:reps] #the first 10 unrelated
GUGN.est <- est.r[180,183,1:reps]
Farthing.est <- est.r[74:149,150,1:reps]
IAC.est <- est.r[151:176,177,1:reps]
## statistic for common relationships
## PO: parent-offspring, FS: fullsibs, HS: halfsibs
## GPGO: grandparent-grandoffspring, UN: uncle-nephew
## GUGN: granduncle-grandnephew, UR: unrelated (first 10 ancestrals)
## RMSE: root means square error
stat.PolyRelatedness$RMSE.PO[i] = sqrt(mean(PO.est-PO.obs)^2)
stat.PolyRelatedness$RMSE.FS[i] = sqrt(mean(FS.est-FS.obs)^2)
stat.PolyRelatedness$RMSE.HS[i] = sqrt(mean(HS.est-HS.obs)^2)
stat.PolyRelatedness$RMSE.GPGO[i] = sqrt(mean(GPGO.est-GPGO.obs)^2)
stat.PolyRelatedness$RMSE.UN[i] = sqrt(mean(UN.est-UN.obs)^2)
stat.PolyRelatedness$RMSE.GUGN[i] = sqrt(mean(GUGN.est-GUGN.obs)^2)
stat.PolyRelatedness$RMSE.UR[i] = sqrt(mean(UR.est)^2) #observed is 0
## IC: Interval of confidence (how many values are within +- of the observed value)
stat.PolyRelatedness$IC.PO[i] = mean(abs(PO.est-PO.obs)<0.05)
stat.PolyRelatedness$IC.FS[i] = mean(abs(FS.est-FS.obs)<0.05)
stat.PolyRelatedness$IC.HS[i] = mean(abs(HS.est-HS.obs)<0.05)
stat.PolyRelatedness$IC.GPGO[i] = mean(abs(GPGO.est-GPGO.obs)<0.05)
stat.PolyRelatedness$IC.UN[i] = mean(abs(UN.est-UN.obs)<0.05)
stat.PolyRelatedness$IC.GUGN[i] = mean(abs(GUGN.est-GUGN.obs)<0.05)
stat.PolyRelatedness$IC.UR[i] = mean(abs(UR.est)<0.05)
## Pearson's correlation for each pedigree (Farthing and IAC)
stat.PolyRelatedness$meancor.Farthing[i] = cor(as.vector(Farthing.obs),
as.vector(Farthing.est),method="pearson")
stat.PolyRelatedness$meancor.IAC[i] = cor(as.vector(IAC.obs),
as.vector(IAC.est),method="pearson")
## RMSE for each pedigree (Farthing and IAC)
stat.PolyRelatedness$RMSE.Farthing[i] = sqrt(mean((as.vector(Farthing.obs) -
as.vector(Farthing.est))^2))
stat.PolyRelatedness$RMSE.IAC[i] = sqrt(mean((as.vector(IAC.obs) -
as.vector(IAC.est))^2))
## Lin's CCC for each pedigree (Farthing and IAC)
stat.PolyRelatedness$CCC.Farthing[i] = as.numeric(CCC(as.vector(Farthing.obs),
as.vector(Farthing.est))$rho.c[1])
stat.PolyRelatedness$CCC.lwrci.Farthing[i] = as.numeric(CCC(as.vector(Farthing.obs),
as.vector(Farthing.est))$rho.c[2])
stat.PolyRelatedness$CCC.uprci.Farthing[i] = as.numeric(CCC(as.vector(Farthing.obs),
as.vector(Farthing.est))$rho.c[3])
stat.PolyRelatedness$CCC.IAC[i] = as.numeric(CCC(as.vector(IAC.obs),
as.vector(IAC.est))$rho.c[1])
stat.PolyRelatedness$CCC.lwrci.IAC[i] = as.numeric(CCC(as.vector(IAC.obs),
as.vector(IAC.est))$rho.c[2])
stat.PolyRelatedness$CCC.uprci.IAC[i] = as.numeric(CCC(as.vector(IAC.obs),
as.vector(IAC.est))$rho.c[3])
}
## Extracting results and computing metrics for AGHmatrix metrics
## "VR","PD","FA"
for(i in 1:nrow(stat.AGHmatrix)){
print(paste0(i,"/",nrow(stat.AGHmatrix)))
## extract the AGHmatrix results
mat.name = load(paste0("biallelic_results/",stat.AGHmatrix$ploidy[i],"/",
stat.AGHmatrix$methods[i],"_loci",stat.AGHmatrix$loci[i],".Rdata"))
est.r = get(mat.name)
rm(mat.name)
ploidy = as.character(stat.AGHmatrix$ploidy[i])
## gather observed relatedness (r)
PO.obs <- obs.r[[ploidy]][181,182,1:reps]
FS.obs <- obs.r[[ploidy]][179,180,1:reps]
HS.obs <- obs.r[[ploidy]][180,181,1:reps]
GPGO.obs <- obs.r[[ploidy]][181,183,1:reps]
UN.obs <- obs.r[[ploidy]][180,182,1:reps]
GUGN.obs <- obs.r[[ploidy]][180,183,1:reps]
Farthing.obs <- obs.r[[ploidy]][74:149,150,1:reps]
IAC.obs <- obs.r[[ploidy]][151:176,177,1:reps]
## gather estimated relatedness (r)
PO.est <- est.r[181,182,1:reps]
FS.est <- est.r[179,180,1:reps]
HS.est <- est.r[180,181,1:reps]
GPGO.est <- est.r[181,183,1:reps]
UN.est <- est.r[180,182,1:reps]
UR.est <- est.r[1:10,1:10,1:reps] #the first 10 unrelated
GUGN.est <- est.r[180,183,1:reps]
Farthing.est <- est.r[74:149,150,1:reps]
IAC.est <- est.r[151:176,177,1:reps]
## statistic for common relationships
## PO: parent-offspring, FS: fullsibs, HS: halfsibs
## GPGO: grandparent-grandoffspring, UN: uncle-nephew
## GUGN: granduncle-grandnephew, UR: unrelated (first 10 ancestrals)
## RMSE: root means square error
stat.AGHmatrix$RMSE.PO[i] = sqrt(mean(PO.est-PO.obs)^2)
stat.AGHmatrix$RMSE.FS[i] = sqrt(mean(FS.est-FS.obs)^2)
stat.AGHmatrix$RMSE.HS[i] = sqrt(mean(HS.est-HS.obs)^2)
stat.AGHmatrix$RMSE.GPGO[i] = sqrt(mean(GPGO.est-GPGO.obs)^2)
stat.AGHmatrix$RMSE.UN[i] = sqrt(mean(UN.est-UN.obs)^2)
stat.AGHmatrix$RMSE.GUGN[i] = sqrt(mean(GUGN.est-GUGN.obs)^2)
stat.AGHmatrix$RMSE.UR[i] = sqrt(mean(UR.est)^2) #observed is 0
## IC: Interval of confidence (how many values are within +- of the observed value)
stat.AGHmatrix$IC.PO[i] = mean(abs(PO.est-PO.obs)<0.05)
stat.AGHmatrix$IC.FS[i] = mean(abs(FS.est-FS.obs)<0.05)
stat.AGHmatrix$IC.HS[i] = mean(abs(HS.est-HS.obs)<0.05)
stat.AGHmatrix$IC.GPGO[i] = mean(abs(GPGO.est-GPGO.obs)<0.05)
stat.AGHmatrix$IC.UN[i] = mean(abs(UN.est-UN.obs)<0.05)
stat.AGHmatrix$IC.GUGN[i] = mean(abs(GUGN.est-GUGN.obs)<0.05)
stat.AGHmatrix$IC.UR[i] = mean(abs(UR.est)<0.05)
## Pearson's correlation for each pedigree (Farthing and IAC)
stat.AGHmatrix$meancor.Farthing[i] = cor(as.vector(Farthing.obs),
as.vector(Farthing.est),method="pearson")
stat.AGHmatrix$meancor.IAC[i] = cor(as.vector(IAC.obs),
as.vector(IAC.est),method="pearson")
## RMSE for each pedigree (Farthing and IAC)
stat.AGHmatrix$RMSE.Farthing[i] = sqrt(mean((as.vector(Farthing.obs) -
as.vector(Farthing.est))^2))
stat.AGHmatrix$RMSE.IAC[i] = sqrt(mean((as.vector(IAC.obs) -
as.vector(IAC.est))^2))
## Lin's CCC for each pedigree (Farthing and IAC)
stat.AGHmatrix$CCC.Farthing[i] = as.numeric(CCC(as.vector(Farthing.obs),
as.vector(Farthing.est))$rho.c[1])
stat.AGHmatrix$CCC.lwrci.Farthing[i] = as.numeric(CCC(as.vector(Farthing.obs),
as.vector(Farthing.est))$rho.c[2])
stat.AGHmatrix$CCC.uprci.Farthing[i] = as.numeric(CCC(as.vector(Farthing.obs),
as.vector(Farthing.est))$rho.c[3])
stat.AGHmatrix$CCC.IAC[i] = as.numeric(CCC(as.vector(IAC.obs),
as.vector(IAC.est))$rho.c[1])
stat.AGHmatrix$CCC.lwrci.IAC[i] = as.numeric(CCC(as.vector(IAC.obs),
as.vector(IAC.est))$rho.c[2])
stat.AGHmatrix$CCC.uprci.IAC[i] = as.numeric(CCC(as.vector(IAC.obs),
as.vector(IAC.est))$rho.c[3])
}
## merging results in a single data.frame and exporting in a csv
stats = rbind(stat.PolyRelatedness,stat.AGHmatrix)
write.csv(stats,"StatResults.csv",quote=FALSE,row.names=FALSE)