Merge pull request #34 from pranavanba/main

Minor Update

config/config.yml

@@ -22,3 +22,8 @@ staging:
 prod:
   inherits: staging
   synFolderID: syn51406700
+
+internal:
+  inherits: staging
+  parquetDirID: syn63135213
+  s3basekey: staging/2024-09-10

pipeline/run-pipeline.R

@@ -2,7 +2,7 @@ tictoc::tic(msg = "INFO: Total execution time")
 # Get config variables
 list2env(
   x = config::get(file = "config/config.yml", 
-                  config = "staging"),
+                  config = "internal"),
   envir = .GlobalEnv
 )
 

scripts/process-data/fitbitactivitylogs.R

@@ -49,6 +49,7 @@ for (col_name in names(df_filtered)) {
 }
 
 # Pivot data frame from long to wide
+cat("recoverutils::melt_df()....")
 df_melted_filtered <- 
   df_filtered %>% 
   recoverutils::melt_df(excluded_concepts = excluded_concepts) %>% 
@@ -58,25 +59,27 @@ df_melted_filtered <-
          if("value" %in% colnames(.)) "value") %>% 
   tidyr::drop_na("value") %>% 
   mutate(value = as.numeric(value))
-cat("recoverutils::melt_df() completed.\n")
+cat("OK\n")
 
 # Generate i2b2 summaries
+cat("recoverutils::stat_summarize()....")
 df_summarized <- 
   df_melted_filtered %>% 
   select(all_of(c("participantidentifier", "startdate", "enddate", "concept", "value"))) %>% 
   recoverutils::stat_summarize() %>% 
   distinct()
-cat("recoverutils::stat_summarize() completed.\n")
+cat("OK\n")
 
 # Add i2b2 columns from concept map (ontology file) and clean the output
+cat("recoverutils::process_df()....")
 output_concepts <- 
   recoverutils::process_df(df_summarized, concept_map, concept_replacements_reversed, concept_map_concepts = "CONCEPT_CD", concept_map_units = "UNITS_CD") %>% 
   dplyr::mutate(nval_num = signif(nval_num, 9)) %>% 
   dplyr::arrange(concept) %>% 
   dplyr::mutate(dplyr::across(.cols = dplyr::everything(), .fns = as.character)) %>% 
   replace(is.na(.), "<null>") %>% 
   dplyr::filter(nval_num != "<null>" | tval_char != "<null>")
-cat("recoverutils::process_df() completed.\n")
+cat("OK\n")
 
 # Identify the participants who have output concepts derived from fitbit variables
 fitbit_participants <- 

scripts/process-data/fitbitdailydata.R

@@ -54,6 +54,7 @@ for (col_name in names(df_filtered)) {
 }
 
 # Pivot data frame from long to wide
+cat("recoverutils::melt_df()....")
 df_melted_filtered <- 
   df_filtered %>% 
   recoverutils::melt_df(excluded_concepts = excluded_concepts) %>% 
@@ -63,19 +64,21 @@ df_melted_filtered <-
          if("value" %in% colnames(.)) "value") %>% 
   tidyr::drop_na("value") %>% 
   mutate(value = as.numeric(value))
-cat("recoverutils::melt_df() completed.\n")
+cat("OK\n")
 
 # Generate i2b2 summaries
+cat("recoverutils::stat_summarize()....")
 df_summarized <- 
   df_melted_filtered %>% 
   rename(startdate = dplyr::any_of(c("date", "datetime"))) %>% 
   mutate(enddate = if (!("enddate" %in% names(.))) NA else enddate) %>% 
   select(all_of(c("participantidentifier", "startdate", "enddate", "concept", "value"))) %>% 
   recoverutils::stat_summarize() %>% 
   distinct()
-cat("recoverutils::stat_summarize() completed.\n")
+cat("OK\n")
 
 # Add i2b2 columns from concept map (ontology file) and clean the output
+cat("recoverutils::process_df()....")
 output_concepts <- 
   recoverutils::process_df(df_summarized, 
              concept_map, 
@@ -87,7 +90,7 @@ output_concepts <-
   dplyr::mutate(dplyr::across(.cols = dplyr::everything(), .fns = as.character)) %>% 
   replace(is.na(.), "<null>") %>% 
   dplyr::filter(nval_num != "<null>" | tval_char != "<null>")
-cat("recoverutils::process_df() completed.\n")
+cat("OK\n")
 
 # Identify the participants who have output concepts derived from fitbit variables
 curr_fitbit_participants <- 

scripts/process-data/fitbitecg.R

@@ -15,7 +15,6 @@
 #' `participantidentifier` and `concept` for all data, and 
 #' `participantidentifier`, `concept`, `year`, `week` for weekly summaries).
 ecg_stat_summarize <- function(df) {
-  cat("Running ecg_stat_summarize()...\n")
 
   if (!is.data.frame(df)) stop("df must be a data frame")
   if (!all(c("participantidentifier", "concept", "value") %in% colnames(df))) stop("'participantidentifier', 'concept', and 'value' columns must be present in df")
@@ -131,6 +130,7 @@ approved_concepts_summarized <-
   )
 
 # Pivot data frame from long to wide
+cat("recoverutils::melt_df()....")
 df_melted_filtered <- 
   df %>% 
   mutate("SinusRhythm" = if_else(resultclassification == "Normal Sinus Rhythm", 1, NA),
@@ -143,9 +143,10 @@ df_melted_filtered <-
          if("value" %in% colnames(.)) "value") %>% 
   tidyr::drop_na("value") %>% 
   mutate(value = as.numeric(value))
-cat("recoverutils::melt_df() completed.\n")
+cat("OK\n")
 
 # Generate i2b2 summaries
+cat("ecg_stat_summarize()....")
 df_summarized <- 
   df_melted_filtered %>% 
   rename(startdate = dplyr::any_of(c("date", "datetime"))) %>% 
@@ -154,9 +155,10 @@ df_summarized <-
   ecg_stat_summarize() %>% 
   mutate(value = as.numeric(value)) %>% 
   distinct()
-cat("ecg_stat_summarize() completed.\n")
+cat("OK\n")
 
 # Add i2b2 columns from concept map (ontology file) and clean the output
+cat("recoverutils::process_df()....")
 output_concepts <- 
   recoverutils::process_df(df_summarized, 
                                 concept_map, 
@@ -168,7 +170,7 @@ output_concepts <-
   dplyr::mutate(dplyr::across(.cols = dplyr::everything(), .fns = as.character)) %>% 
   replace(is.na(.), "<null>") %>% 
   dplyr::filter(nval_num != "<null>" | tval_char != "<null>")
-cat("recoverutils::process_df() completed.\n")
+cat("OK\n")
 
 # Identify the participants who have output concepts derived from fitbit variables
 curr_fitbit_participants <- 

scripts/process-data/fitbitintradaycombined.R

@@ -72,6 +72,7 @@ for (col_name in names(df_filtered)) {
 }
 
 # Pivot data frame from long to wide
+cat("recoverutils::melt_df()....")
 df_melted_filtered <- 
   df_filtered %>% 
   recoverutils::melt_df(excluded_concepts = excluded_concepts) %>% 
@@ -81,27 +82,29 @@ df_melted_filtered <-
          if("value" %in% colnames(.)) "value") %>% 
   tidyr::drop_na("value") %>% 
   mutate(value = as.numeric(value))
-cat("recoverutils::melt_df() completed.\n")
+cat("OK\n")
 
 # Generate i2b2 summaries
+cat("recoverutils::stat_summarize()....")
 df_summarized <- 
   df_melted_filtered %>% 
   rename(startdate = dplyr::any_of(c("date", "datetime"))) %>% 
   mutate(enddate = if (!("enddate" %in% names(.))) NA else enddate) %>% 
   select(all_of(c("participantidentifier", "startdate", "enddate", "concept", "value"))) %>% 
   recoverutils::stat_summarize() %>% 
   distinct()
-cat("recoverutils::stat_summarize() completed.\n")
+cat("OK\n")
 
 # Add i2b2 columns from concept map (ontology file) and clean the output
+cat("recoverutils::process_df()....")
 output_concepts <- 
   recoverutils::process_df(df_summarized, concept_map, concept_replacements_reversed, concept_map_concepts = "CONCEPT_CD", concept_map_units = "UNITS_CD") %>% 
   dplyr::mutate(nval_num = signif(nval_num, 9)) %>% 
   dplyr::arrange(concept) %>% 
   dplyr::mutate(dplyr::across(.cols = dplyr::everything(), .fns = as.character)) %>% 
   replace(is.na(.), "<null>") %>% 
   dplyr::filter(nval_num != "<null>" | tval_char != "<null>")
-cat("recoverutils::process_df() completed.\n")
+cat("OK\n")
 
 # Identify the participants who have output concepts derived from fitbit variables
 curr_fitbit_participants <- 

scripts/process-data/fitbitsleeplogs.R

@@ -15,7 +15,6 @@
 #' `participantidentifier` and `concept` for all data, and 
 #' `participantidentifier`, `concept`, `year`, `week` for weekly summaries).
 sleeplogs_stat_summarize <- function(df) {
-  cat("Running sleeplogs_stat_summarize()...\n")
 
   if (!is.data.frame(df)) stop("df must be a data frame")
   if (!all(c("participantidentifier", "concept", "value") %in% colnames(df))) stop("'participantidentifier', 'concept', and 'value' columns must be present in df")
@@ -332,6 +331,7 @@ for (col_name in names(df_filtered)) {
 }
 
 # Pivot data frames from long to wide
+cat("recoverutils::melt_df()....")
 df_melted_filtered <- 
   df_filtered %>% 
   recoverutils::melt_df(excluded_concepts = excluded_concepts) %>% 
@@ -366,9 +366,10 @@ numepisodes_df_melted_filtered_weekly <-
   tidyr::drop_na("value") %>%
   mutate(value = as.numeric(value))
 
-cat("recoverutils::melt_df() completed.\n")
+cat("OK\n")
 
 # Generate i2b2 summaries
+cat("sleeplogs_stat_summarize()....")
 df_summarized <- 
   df_melted_filtered %>% 
   select(all_of(c("participantidentifier", "startdate", "enddate", "concept", "value"))) %>% 
@@ -397,17 +398,18 @@ final_df_summarized <-
                    numepisodes_df_summarized_weekly) %>% 
   dplyr::distinct()
 
-cat("sleeplogs_stat_summarize() completed.\n")
+cat("OK\n")
 
 # Add i2b2 columns from concept map (ontology file) and clean the output
+cat("recoverutils::process_df()....")
 output_concepts <- 
   recoverutils::process_df(final_df_summarized, concept_map, concept_replacements_reversed, concept_map_concepts = "CONCEPT_CD", concept_map_units = "UNITS_CD") %>% 
   dplyr::mutate(nval_num = signif(nval_num, 9)) %>% 
   dplyr::arrange(concept) %>% 
   dplyr::mutate(dplyr::across(.cols = dplyr::everything(), .fns = as.character)) %>% 
   replace(is.na(.), "<null>") %>% 
   dplyr::filter(nval_num != "<null>" | tval_char != "<null>")
-cat("recoverutils::process_df() completed.\n")
+cat("OK\n")
 
 # Identify the participants who have output concepts derived from fitbit variables
 curr_fitbit_participants <- 

scripts/process-data/healthkitv2electrocardiogram.R

@@ -15,7 +15,6 @@
 #' `participantidentifier` and `concept` for all data, and 
 #' `participantidentifier`, `concept`, `year`, `week` for weekly summaries).
 ecg_stat_summarize <- function(df) {
-  cat("Running ecg_stat_summarize()...\n")
 
   if (!is.data.frame(df)) stop("df must be a data frame")
   if (!all(c("participantidentifier", "concept", "value") %in% colnames(df))) stop("'participantidentifier', 'concept', and 'value' columns must be present in df")
@@ -136,6 +135,7 @@ approved_concepts_summarized <-
   )
 
 # Pivot data frame from long to wide
+cat("recoverutils::melt_df()....")
 df_melted_filtered <- 
   df %>% 
   mutate("SinusRhythm" = if_else(classification == "SinusRhythm", 1, NA),
@@ -148,18 +148,20 @@ df_melted_filtered <-
          if("value" %in% colnames(.)) "value") %>% 
   tidyr::drop_na("value") %>% 
   mutate(value = as.numeric(value))
-cat("recoverutils::melt_df() completed.\n")
+cat("OK\n")
 
 # Generate i2b2 summaries
+cat("ecg_stat_summarize()....")
 df_summarized <- 
   df_melted_filtered %>% 
   select(all_of(c("participantidentifier", "startdate", "enddate", "concept", "value"))) %>% 
   ecg_stat_summarize() %>% 
   mutate(value = as.numeric(value)) %>% 
   distinct()
-cat("ecg_stat_summarize() completed.\n")
+cat("OK\n")
 
 # Add i2b2 columns from concept map (ontology file) and clean the output
+cat("recoverutils::process_df()....")
 output_concepts <- 
   recoverutils::process_df(df_summarized, 
                                 concept_map, 
@@ -171,7 +173,7 @@ output_concepts <-
   dplyr::mutate(dplyr::across(.cols = dplyr::everything(), .fns = as.character)) %>% 
   replace(is.na(.), "<null>") %>% 
   dplyr::filter(nval_num != "<null>" | tval_char != "<null>")
-cat("recoverutils::process_df() completed.\n")
+cat("OK\n")
 
 # Identify the participants who have output concepts derived from healthkit variables
 curr_hk_participants <- 

scripts/process-data/healthkitv2samples.R

@@ -74,20 +74,22 @@ df_melted_filtered <-
 cat("Melt and filtering step completed.\n")
 
 # Generate i2b2 summaries
+cat("recoverutils::stat_summarize()....")
 df_summarized <- 
   df_melted_filtered %>% 
   rename(enddate = "date") %>% 
   select(all_of(c("participantidentifier", "startdate", "enddate", "concept", "value"))) %>% 
   recoverutils::stat_summarize() %>% 
   distinct()
-cat("recoverutils::stat_summarize() completed.\n")
+cat("OK\n")
 
 tmp_concept_replacements <- c("respiratoryrate" = "breathingrate",
                               "heartratevariability" = "hrv",
                               "heartrate" = "avghr",
                               "oxygensaturation" = "spo2avg")
 
 # Add i2b2 columns from concept map (ontology file) and clean the output
+cat("recoverutils::process_df()....")
 output_concepts <- 
   recoverutils::process_df(df_summarized, 
              concept_map, 
@@ -99,7 +101,7 @@ output_concepts <-
   dplyr::mutate(dplyr::across(.cols = dplyr::everything(), .fns = as.character)) %>% 
   replace(is.na(.), "<null>") %>% 
   dplyr::filter(nval_num != "<null>" | tval_char != "<null>")
-cat("recoverutils::process_df() completed.\n")
+cat("OK\n")
 
 # Identify the participants who have output concepts derived from healthkit variables
 curr_hk_participants <- 

scripts/process-data/healthkitv2statistics.R

@@ -49,17 +49,19 @@ df_melted_filtered <-
 cat("Melt and filtering step completed.\n")
 
 # Generate i2b2 summaries
+cat("recoverutils::stat_summarize()....")
 df_summarized <- 
   df_melted_filtered %>% 
   rename(enddate = "date") %>% 
   select(all_of(c("participantidentifier", "startdate", "enddate", "concept", "value"))) %>% 
   recoverutils::stat_summarize() %>% 
   distinct()
-cat("recoverutils::stat_summarize() completed.\n")
+cat("OK\n")
 
 tmp_concept_replacements <- c("dailysteps" = "steps")
 
 # Add i2b2 columns from concept map (ontology file) and clean the output
+cat("recoverutils::process_df()....")
 output_concepts <- 
   recoverutils::process_df(df_summarized, 
              concept_map, 
@@ -71,7 +73,7 @@ output_concepts <-
   dplyr::mutate(dplyr::across(.cols = dplyr::everything(), .fns = as.character)) %>% 
   replace(is.na(.), "<null>") %>% 
   dplyr::filter(nval_num != "<null>" | tval_char != "<null>")
-cat("recoverutils::process_df() completed.\n")
+cat("OK\n")
 
 # Identify the participants who have output concepts derived from healthkit variables
 curr_hk_participants <- 

scripts/process-data/participant_devices.R

@@ -66,6 +66,7 @@ df_joined <-
 # Add i2b2 columns from concept map (ontology file) and clean the output
 concept_map_concepts <- "CONCEPT_CD"
 concept_map_units <- "UNITS_CD"
+cat("recoverutils::process_df()....")
 output_concepts <- 
   df_joined %>% 
   dplyr::mutate(valtype_cd = dplyr::case_when(class(value) == "numeric" ~ "N", 
@@ -84,7 +85,7 @@ output_concepts <-
   dplyr::select(participantidentifier, startdate, enddate, 
                 concept, valtype_cd, nval_num, tval_char, UNITS_CD) %>% 
   dplyr::rename(units_cd = UNITS_CD)
-cat("recoverutils::process_df() completed.\n")
+cat("OK\n")
 
 # Write the output
 output_concepts %>%