-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy path25_age_stratification_annotate.R
232 lines (155 loc) · 8.33 KB
/
25_age_stratification_annotate.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
#' Title
#' Per age group, it creates the split reads annotation files
#' @param age.groups
#' @param project.name
#' @param gtf.version
#' @param project.id
#' @param age.samples.clusters
#' @param results.folder
#' @param supportive.reads
#'
#' @return
#' @export
#'
#' @examples
age_stratification_annotate <- function (age.groups,
project.name,
gtf.version,
project.id,
age.samples.clusters,
results.folder) {
all_samples_used <- NULL
local_results_folder <- file.path(results.folder, "/base_data/")
## Loop per age cluster
doParallel::registerDoParallel(3)
all_samples_age_used <- foreach(i = seq(length(age.groups))) %dopar%{
age_cluster <- age.groups[i]
# message(age_cluster)
# age_cluster <- age.groups[1]
# age_cluster <- age.groups[2]
## THE AGE STRATIFICATION SHOULD BE DONE BY BODY SITE, AND STORED IN THE PATH OF THE BODY SITE.
split_read_counts_age <- NULL
all_split_reads_age <- NULL
samples_age <- NULL
j <- 1
## GENERATE THE COUNTS AND ANNOTATED SPLIT READ COUNTS BY AGE CLUSTER ---------------------------------------------------
for ( cluster_id in (age.samples.clusters$cluster %>% unique()) ) {
# cluster_id <- (age.samples.clusters$cluster %>% unique())[2]
# print(paste0(Sys.time(), " - ", cluster_id, ", ", age_cluster))
## Get the samples from the current tissue according to the current age cluster
samples <- age.samples.clusters %>%
filter(age_group == age_cluster, cluster == cluster_id) %>%
pull(individual)
if ( samples %>% length() > 0 ) {
samples_age <- c(samples_age, samples)
print(paste0(cluster_id, " - ", age_cluster, ": ", samples %>% length(), " samples... "))
## Load the annotated split reads (the splicing project contains all samples QC'ed and checked with recount3 origin)
all_split_reads_local <- readRDS(file = paste0(local_results_folder, "/", project.id,
"_", cluster_id, "_all_split_reads.rds"))
# Load split read counts
split_read_counts_local <- readRDS(file = paste0(local_results_folder, "/", project.id,
"_", cluster_id, "_split_read_counts.rds"))
if ( is.null(names(split_read_counts_local)) ) {
split_read_counts_local <- split_read_counts_local %>%
as_tibble(rownames = "junID")
}
## Only split read counts from current cluster
split_read_counts_local <- split_read_counts_local %>%
dplyr::select(c("junID", any_of(samples)))
# any((split_read_counts_local %>% names) == "GTEX-1399S-1426-SM-5PNYQ.1")
## Only split read counts with at least 1 supportive reads after filtering using the local cluster of samples
split_read_counts_local <- split_read_counts_local %>%
mutate(total = rowSums(select_if(., is.numeric))) %>%
filter(total >= 1) %>%
dplyr::select(-total)
## Only split reads selected
all_split_reads_local <- all_split_reads_local %>%
filter(junID %in% split_read_counts_local$junID)
if ( !identical(all_split_reads_local$junID,
split_read_counts_local$junID) ) {
logger::log_info("ERROR!")
break;
}
if ( j == 1 ) {
split_read_counts_age <- split_read_counts_local
all_split_reads_age <- all_split_reads_local
j <- 2
} else {
## The age stratification analysis is done per body site category,
## thus, only annotated introns overlapping the clusters from each
## body site are considered
all_split_reads_age <- data.table::rbindlist(list(all_split_reads_local, all_split_reads_age)) %>%
distinct(junID, .keep_all = T)
# split_read_counts_local$junID %>% unique() %>% length()
# split_read_counts_age$junID %>% unique() %>% length()
#
# intersect(split_read_counts_local$junID, split_read_counts_age$junID) %>% unique() %>% length() +
# setdiff(split_read_counts_local$junID, split_read_counts_age$junID) %>% unique() %>% length() +
# setdiff(split_read_counts_age$junID, split_read_counts_local$junID) %>% unique() %>% length()
## OUTTER JOIN
split_read_counts_age <- merge(x = split_read_counts_local,
y = split_read_counts_age,
by = "junID",
all = TRUE)
# split_read_counts_age_temp %>% as_tibble()
# split_read_counts_age_temp[1,] %>% as.data.frame
}
#print(paste0(split_read_counts_age %>% names() %>% length(), " samples processed in total."))
rm(all_split_reads_local)
rm(split_read_counts_local)
rm(samples)
gc()
}
}
# split_read_counts_age %>% names() %>% length()
# split_read_counts_age[(split_read_counts_age[,2] > 0),]
# split_read_counts_age[1,] %>% as.data.frame()
split_read_counts_age %>%
as_tibble()
logger::log_info(" - SAVING RESULTS ... ")
split_read_counts_age[is.na(split_read_counts_age)] <- 0
split_read_counts_age <- split_read_counts_age %>%
mutate(total = rowSums(select_if(., is.numeric),na.rm = TRUE)) %>%
filter(total >= 1) %>%
dplyr::select(-total)
# split_read_counts_age <- split_read_counts_age[rowSums(is.na(split_read_counts_age[,2:ncol(split_read_counts_age)])) !=
# (ncol(split_read_counts_age)-1),]
# split_read_counts_age %>%
# dplyr::select(c("junID", "GTEX-1399S-1426-SM-5PNYQ.1")) %>%
# as_tibble() %>%
# mutate(total = rowSums(select_if(., is.numeric))) %>%
# filter(total >= 1)
all_split_reads_age <- all_split_reads_age %>%
distinct(junID, .keep_all = T) %>%
filter( junID %in% (split_read_counts_age$junID %>% unique()) )
############################################
## SAVE RESULTS FOR THE CURRENT AGE CLUSTER
############################################
saveRDS(object = samples_age,
file = paste0(local_results_folder, "/", project.id, "_", age_cluster, "_samples_used.rds"))
## Split read counts for age
saveRDS(object = split_read_counts_age %>% data.table::as.data.table(),
file = paste0(local_results_folder, "/", project.id, "_", age_cluster, "_split_read_counts.rds"))
## All split reads for age
saveRDS(object = all_split_reads_age %>% data.table::as.data.table(),
file = paste0(local_results_folder, "/", project.id, "_", age_cluster, "_all_split_reads.rds"))
logger::log_info(age_cluster, " - ", project.id, " - RESULTS SAVED!")
return(samples_age)
}
all_samples_age_used %>% head()
saveRDS(object = all_samples_age_used %>% unlist,
file = paste0(local_results_folder, "/", project.id, "_age_all_samples_used.rds"))
metadata_project <- readRDS(file = paste0(local_results_folder, "/", project.id, "_samples_metadata.rds") )
## Save age-related metadata
saveRDS(object = metadata_project %>%
filter(sample_id %in% (all_samples_age_used %>% unlist)) %>%
inner_join(y = age.samples.clusters %>% dplyr::select(individual, age_group),
by = c("sample_id" = "individual")) %>%
dplyr::select(-cluster) %>%
dplyr::rename(cluster = age_group),
file = paste0(local_results_folder, "/", project.id, "_age_samples_metadata.rds"))
## Save age-related clusters
saveRDS(object = age.groups %>% unique(),
file = paste0(local_results_folder, "/", project.id, "_age_clusters_used.rds"))
gc()
}