Rewrite

bin/makeMetadata_dauer.R

@@ -0,0 +1,72 @@
+#!/usr/bin/env -S Rscript --vanilla
+library(dplyr)
+library(tidyr)
+library(tibble)
+library(stringr)
+library(readr)
+library(purrr)
+library(data.table)
+
+#==============================================================================#
+# Arguments
+#==============================================================================#
+# 1 - A list of input files
+# 2 - the path to the well_mask - HARDCODE NOW
+# 3 - the group argument from main.nf - default is "plate,well"
+# 4 - the edited pipeline path
+# 5 - the out path
+args <- commandArgs(trailingOnly = TRUE)
+#==============================================================================#
+# Make Metadata NEEDS TO BE ADAPATBLE TO MULTIPLE WAVELENGTHS
+#==============================================================================#
+# parse file names from directory - need wavelength in file name
+meta1 <- read_delim(
+  args[1], 
+  col_names = FALSE, 
+  delim = "\t") %>% 
+  select(file_path = X1) %>% 
+  extract(file_path, into = "file", remove = FALSE, regex = ".*/(.*)$") %>% 
+  extract(file, 
+          remove = FALSE, 
+          regex = "^(.*)-(.*)-(.*)-(.*)_(.*)_(.*)\\.(.*)$", 
+          into = c("date","exp","plate","mag","well","wave","TIF")) %>% 
+  select(-TIF) %>%
+  dplyr::mutate(row = stringr::str_extract(well, pattern = "[A-Z]"),
+                col = stringr::str_extract(well, pattern = "[0-9][0-9]"),
+                Image_PathName_wellmask_98.png = stringr::str_replace(args[2], pattern = "([^/]+$)", replacement = ""),
+                Image_FileName_wellmask_98.png = stringr::str_extract(args[2], pattern = "([^/]+$)"))
+
+# add group
+groups <- stringr::str_split(args[3], pattern = ",")[[1]]
+meta1$group <- apply( meta1[, groups], 1, paste, collapse = "_")
+
+# add image types and set metadata names - hardcode image names - needs to be flexible for multiple pipeline profiles
+meta2 <- meta1 %>%
+  tidyr::pivot_wider(names_from = wave, values_from = c(file, file_path), ) %>%
+  dplyr::rename(Image_FileName_RawBF = file_w1,
+                Image_PathName_RawBF = file_path_w1,  
+                Image_FileName_RawRFP = file_w2,
+                Image_PathName_RawRFP = file_path_w2) %>%
+  dplyr::mutate(
+    Image_PathName_RawRFP = stringr::str_replace(
+      Image_PathName_RawRFP, pattern = "([^/]+$)", replacement = ""
+    ),
+    Image_PathName_RawBF = stringr::str_replace(
+      Image_PathName_RawBF, pattern = "([^/]+$)", replacement = ""
+    )
+  ) %>%  
+  dplyr::select(
+    Metadata_Experiment = exp,
+    Metadata_Date = date,
+    Metadata_Plate = plate,
+    Metadata_Well = well,
+    Metadata_Group = group,
+    Metadata_Magnification = mag,
+    Image_FileName_RawBF,
+    Image_PathName_RawBF,
+    Image_FileName_RawRFP,
+    Image_PathName_RawRFP,
+    Image_FileName_wellmask_98.png,
+    Image_PathName_wellmask_98.png)
+
+write.table(meta2, file = "metadata.csv", quote=FALSE, sep=',', row.names = F)

bin/makeMetadata_toxin.R

@@ -0,0 +1,58 @@
+#!/usr/bin/env -S Rscript --vanilla
+library(dplyr)
+library(tidyr)
+library(tibble)
+library(stringr)
+library(readr)
+library(purrr)
+library(data.table)
+
+#==============================================================================#
+# Arguments
+#==============================================================================#
+# 1 - A list of input files
+# 2 - the path to the well_mask - HARDCODE NOW
+# 3 - the group argument from main.nf - default is "plate,well"
+# 4 - the edited pipeline path
+# 5 - the out path
+args <- commandArgs(trailingOnly = TRUE)
+
+#==============================================================================#
+# Make Metadata NEEDS TO BE ADAPATBLE TO MULTIPLE WAVELENGTHS
+#==============================================================================#
+# parse file names from directory - need wavelength in file name
+meta1 <- read_delim(
+  args[1], 
+  col_names = FALSE, 
+  delim = "\t") %>% 
+  select(file_path = X1) %>% 
+  extract(file_path, into = "file", remove = FALSE, regex = ".*/(.*)$") %>% 
+  extract(file, 
+          remove = FALSE, 
+          regex = "^(.*)-(.*)-(.*)-(.*)_(.*)\\.(.*)$", 
+          into = c("date","exp","plate","mag","well","TIF")) %>% 
+  select(-TIF) %>%
+  dplyr::mutate(row = stringr::str_extract(well, pattern = "[A-Z]"),
+                col = stringr::str_extract(well, pattern = "[0-9][0-9]"),
+                Image_PathName_wellmask_98.png = stringr::str_replace(args[2], pattern = "([^/]+$)", replacement = ""),
+                Image_FileName_wellmask_98.png = stringr::str_extract(args[2], pattern = "([^/]+$)"))
+
+# add group
+groups <- stringr::str_split(args[3], pattern = ",")[[1]]
+meta1$group <- apply( meta1[, groups], 1, paste, collapse = "_")
+
+# add image types and set metadata names - hardcode image names - needs to be flexible for multiple pipeline profiles
+meta2 <- meta1 %>%
+  dplyr::mutate(Image_PathName_RawBF = stringr::str_replace(file_path, pattern = "([^/]+$)", replacement = "")) %>%
+  dplyr::select(Metadata_Experiment = exp,
+                Metadata_Date = date,
+                Metadata_Plate = plate,
+                Metadata_Well = well,
+                Metadata_Group = group,
+                Metadata_Magnification = mag,
+                Image_FileName_RawBF = file,
+                Image_PathName_RawBF,
+                Image_FileName_wellmask_98.png,
+                Image_PathName_wellmask_98.png)
+
+write.table(meta2, file = "metadata.csv", quote=FALSE, sep=',', row.names = F)

bin/proc_CP_output_dauer.R → bin/proc_CP_output.R

@@ -1,24 +1,25 @@
-# /usr/bin/Rscript
+#!/usr/bin/env -S Rscript --vanilla
 library(fs)
 library(dplyr)
 library(tidyr)
 library(tibble)
 library(stringr)
 library(readr)
-library(glue)
 library(purrr)
 
 #==============================================================================#
 # Arguments
 #==============================================================================#
 # 1 - out directory path
+# 2 - project name
+# 3 - run stamp
 args <- commandArgs(trailingOnly = TRUE)
 
 #==============================================================================#
 # Read CP output data
 #==============================================================================#
 # get the output for each model
-dir <- glue::glue("{args[1]}/processed_data")
+dir <- "processed_data"
 
 # read in files and manipulate with  
 model_df <- dir %>%
@@ -36,10 +37,7 @@ model_df_list <- split.data.frame(model_df, model_df$model)
 lapply(seq_along(model_df_list), function(i) assign(names(model_df_list)[i], model_df_list[[i]], envir = .GlobalEnv))
 
 # save as R.data
-proj_name <- stringr::str_extract(args[1], pattern = "[^\\/]+(?=(?:\\/[^\\/]+){1}$)")
-run_stamp <- stringr::str_extract(args[1], pattern = "([^/]+$)")
+proj_name <- args[1]
+run_stamp <- args[2]
 save(list = c(ls(pattern = "model.outputs")),
-     file = glue::glue("{args[1]}/processed_data/{proj_name}_{run_stamp}.RData"))
-
-# clean up extra CP_outputs for now
-system(command = glue::glue("if [ -d {args[1]}/CP_output ]; then rm -Rf {args[1]}/CP_output; fi"))
+     file = paste0("processed_data/", args[1], "_", args[2], ".RData"))