-
Notifications
You must be signed in to change notification settings - Fork 1
/
identifiers.R
208 lines (151 loc) · 5.88 KB
/
identifiers.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
# Raw R code chunks for 'identifiers.Rmd'
## @knitr libraries
library("dplyr")
library("tidyr")
library("stringr")
library("readr")
## @knitr sourceCompiling
source("compiling.R")
rm(list = ls(pattern = '_df|gather'))
## @knitr unique_PITs_per_vTag
unique_PITs_Tab <- cleanSites %>%
filter(!is.na(vTagID), !is.na(PIT_Tag)) %>%
group_by(vTagID) %>%
summarize(uniquePITs = length(unique(PIT_Tag))) %>%
select(uniquePITs) %>%
table
unique_PITs_Tab
## @knitr unique_vTags_per_PIT
unique_vTags_Tab <- cleanSites %>%
filter(!is.na(vTagID), !is.na(PIT_Tag)) %>%
group_by(PIT_Tag) %>%
summarize(uniquevTagIDs = length(unique(vTagID))) %>%
select(uniquevTagIDs) %>%
table
unique_vTags_Tab
## @knitr maxUniques
maxUniques <- c(as.numeric(attr(unique_PITs_Tab, 'dimnames')$.),
as.numeric(attr(unique_vTags_Tab, 'dimnames')$.)) %>% max
## @knitr vTag_PIT_lookup_DF
vTagID_PIT <- cleanSites %>%
filter(!is.na(vTagID), !is.na(PIT_Tag)) %>%
distinct(vTagID, PIT_Tag) %>%
select(vTagID, PIT_Tag)
## @knitr vTag_PIT_lookup_Funs
vTagID_to_PITs <- function(in_vTagIDs){
one_vTagID <- function(in_vTagID){
as.character(filter(vTagID_PIT, vTagID == as.integer(in_vTagID))$PIT_Tag)
}
return(sapply(in_vTagIDs, one_vTagID, USE.NAMES = FALSE))
}
PIT_to_vTagID <- function(in_PITs){
one_PIT <- function(in_PIT){
as.integer(filter(vTagID_PIT, PIT_Tag == as.character(in_PIT))$vTagID)
}
return(sapply(in_PITs, one_PIT, USE.NAMES = FALSE))
}
## @knitr newest_vTagID_fun
findNewest_vTagID <- function(focal_vTagIDs, refDF = cleanSites){
new_vTagID <- (refDF %>%
filter(vTagID %in%
as.integer(focal_vTagIDs[!is.na(focal_vTagIDs)])) %>%
arrange(desc(Date)))$vTagID[1]
return(new_vTagID)
}
## @knitr equiv_vTagID_PIT
equiv_vTagID_PIT <-
vTagID_PIT %>%
group_by(PIT_Tag) %>%
# Combining all equivalent vTagIDs into one character column, separated by ':'
summarize(vTagIDs = paste(vTagID, collapse = ':')) %>%
# ... then split them back up
separate(vTagIDs, paste0('vTagID_', seq(maxUniques)), sep = ':',
fill = 'right') %>%
# Remove rows where only one vTagID matches with the PIT_Tag
filter(!is.na(vTagID_2)) %>%
# Convert vTagID columns back to integer for compatibility with `cleanSites` data frame
mutate_each(funs(as.integer), starts_with('vTagID')) %>%
# Using 'standard evaluation' by using `select_`, which allows use of `maxUniques`
select_(.dots = c(paste0('vTagID_', seq(maxUniques)), 'PIT_Tag'))
## @knitr equiv_vTagID
equiv_vTagID <- data_frame(input = as.vector(t(equiv_vTagID_PIT[,1:maxUniques])),
newest = rep(apply(equiv_vTagID_PIT[,1:maxUniques], 1,
findNewest_vTagID),
each = maxUniques)) %>%
filter(!is.na(input))
## @knitr expand_vTagID_PIT_NewRows
vTagID_PIT_NewRows <- equiv_vTagID %>%
# Group by individual
group_by(newest) %>%
# Create character strings of all PIT_Tags and vTagIDs for each individual,
# separated by colons
summarize(PIT_Tags = paste0(vTagID_PIT$PIT_Tag[vTagID_PIT$vTagID %in% input],
collapse = ':'),
vTagIDs = paste0(input, collapse = ':')) %>%
# Separate PIT_Tags into separate columns (used 20 bc it's definitely more than what
# we'll observe)
separate(PIT_Tags, paste0('PIT_Tags_', seq(20)), sep = ':',
fill = 'right') %>%
# Condense into a single column by adding rows, or "lengthening" data frame
gather(PIT_name, PIT_Tag, -newest, -vTagIDs, na.rm = TRUE) %>%
# These columns no longer needed
select(-PIT_name, -newest) %>%
# Do the same as above for the vTagIDs
separate(vTagIDs, paste0('vTagIDs_', seq(10)), sep = ':',
fill = 'right') %>%
gather(vTagID_name, vTagID, -PIT_Tag, na.rm = TRUE) %>%
# Make vTagID integer for compatibility
mutate(vTagID = as.integer(vTagID)) %>%
select(-vTagID_name) %>%
# Filter for unique combinations
distinct(vTagID, PIT_Tag)
## @knitr expand_vTagID_PIT
vTagID_PIT <- bind_rows(vTagID_PIT, vTagID_PIT_NewRows) %>%
distinct(vTagID, PIT_Tag) %>%
arrange(vTagID)
## @knitr max_vTagIDs
max_vTagIDs <- vTagID_PIT %>%
group_by(vTagID) %>%
summarize(len = n()) %>%
select(len) %>%
max
## @knitr PITs_w_vTag
PITs_w_vTag <- vTagID_PIT %>%
group_by(PIT_Tag) %>%
summarize(vTagID_new = findNewest_vTagID(vTagID)) %>%
rename(vTagID = vTagID_new) %>%
mutate(
System = sapply(PIT_Tag, function(x){
tail(cleanSites$System[cleanSites$PIT_Tag == x] %>% na.omit, 1)},
USE.NAMES = FALSE),
Subsystem = sapply(PIT_Tag, function(x){
tail(cleanSites$Subsystem[cleanSites$PIT_Tag == x] %>% na.omit, 1)},
USE.NAMES = FALSE)
) %>%
arrange(System, Subsystem, PIT_Tag) %>%
select(System, Subsystem, PIT_Tag, vTagID)
## @knitr dropped_vTagIDs
dropped_vTagIDs <- equiv_vTagID %>%
filter(! input %in% equiv_vTagID$newest) %>%
rename(dropped = input, new = newest)
## @knitr newestID_master
cleanSites_newIDs <- cleanSites %>%
mutate(
vTagID = sapply(vTagID,
function(x){
ifelse(x %in% dropped_vTagIDs$dropped,
dropped_vTagIDs$new[dropped_vTagIDs$dropped == x],
x)
})
)
## @knitr knownID_master
masterCaps <- list(
# valid vTagIDs:
cleanSites_newIDs %>% filter(!is.na(vTagID)),
# NA vTagIDs, but valid PITs:
cleanSites_newIDs %>%
filter(is.na(vTagID),
PIT_Tag %in% vTagID_PIT$PIT_Tag) %>%
mutate(vTagID = PIT_to_vTagID(PIT_Tag))
) %>%
bind_rows