pushing some final changes to allow new runs'

Author: miloswrath

README.md

@@ -20,3 +20,12 @@
 - [ ] add study rule ->
     - if sub already in obs -> move to int 
     - all need to be after start date 
+
+### adding sleep logs
+- [x] need to return the sleep files from `/sleep` inside the accel folder in RDSS
+- [x] modify `code/utils/comparison_utils` to also add the sleep logs to the linkage, storing the correct study inside
+    - [ ] need to split running by session and split sleep logs by session
+    
+
+
+- [ ] create a new util that transforms the current logs into the GGIR available one

app/.gitkeeper

@@ -0,0 +1,152 @@
+#!/usr/bin/env Rscript
+
+library(tidyr)
+library(plyr)
+library(optparse)
+library(GGIR)
+
+main <- function() {
+  # Define the option list
+  option_list <- list(
+    make_option(c("-p", "--project_dir"), type="character", 
+                default="/mnt/nfs/lss/vosslabhpc/Projects/BOOST/InterventionStudy/3-Experiment/data/act-int-test/",
+                help="Path to the project directory", metavar="character"),
+    make_option(c("-d", "--deriv_dir"), type="character", 
+                default="/derivatives/GGIR-2.8.2-test2/",
+                help="Path to the derivatives directory", metavar="character")
+  )
+
+  # Parse the options
+  opt_parser <- OptionParser(option_list=option_list)
+  opt <- parse_args(opt_parser)
+
+  # Assign variables
+  ProjectDir <- opt$project_dir
+  ProjectDerivDir <- opt$deriv_dir
+
+  # Print values to verify
+  print(paste("Project Directory:", ProjectDir))
+  print(paste("Derivatives Directory:", ProjectDerivDir))
+
+  # Helper functions
+  SubjectGGIRDeriv <- function(x) {
+    a <- dirname(x)
+    paste0(ProjectDir, ProjectDerivDir, a)
+  }
+
+  datadirname <- function(x) {
+    b <- dirname(x)
+    paste0(ProjectDir, b)
+  }
+
+  # Gather subject directories
+  directories <- list.dirs(ProjectDir, recursive = FALSE)
+  subdirs <- directories[grepl("sub-*", directories)]
+  print(paste("subdirs: ", subdirs))
+
+  # Create project-specific derivatives GGIR folder if it doesn't already exist
+  if (!dir.exists(paste0(ProjectDir, ProjectDerivDir))) {
+    dir.create(paste0(ProjectDir, ProjectDerivDir))
+  }
+
+  # List accel.csv files within subject-specific folders
+  filepattern <- "*accel.csv"
+  GGIRfiles <- list.files(subdirs, pattern = filepattern, recursive = TRUE,  
+                          include.dirs = TRUE, full.names = TRUE, no.. = TRUE)
+  print(paste("GGIR Files before splitting: ", GGIRfiles))
+
+  # Split files at the "//" so we only have paths from sub-XXX/ses-.....
+  GGIRfiles <- sapply(strsplit(GGIRfiles, "//", fixed = TRUE), function(x) paste(x[2]))
+
+  print(paste("GGIR Files after splitting: ", GGIRfiles))
+
+
+  # Ensure directory structure exists for GGIR processing
+  for (i in GGIRfiles) {
+    if (!dir.exists(SubjectGGIRDeriv(i))) {
+      dir.create(SubjectGGIRDeriv(i), recursive = TRUE)
+    }
+  }
+
+
+  # Run GGIR loop
+  for (r in GGIRfiles) {
+    if (dir.exists(paste0(SubjectGGIRDeriv(r), "/output_beh"))) {
+      next
+    } else {
+      datadir <- paste0(normalize(path(datadirname(r)))
+      outputdir <- paste0(SubjectGGIRDeriv(r))
+      print(paste("datadir: ", datadir))
+      print(paste("outputdir: ", outputdir))
+      if (!dir.exists(datadir)) {
+        stop(paste("Error: datadir does not exist ->", datadir))
+      }
+    # Force evaluation before calling GGIR
+    assign("datadir", datadir, envir = .GlobalEnv)
+    assign("outputdir", outputdir, envir = .GlobalEnv)
+
+      try(g.shell.GGIR(
+        mode = 1:5,
+        datadir = get("datadir", envir = .GlobalEnv),
+        outputdir = get("outputdir", envir = .GlobalEnv),
+        overwrite = FALSE,
+        print.filename = TRUE,
+        storefolderstructure = FALSE,
+        windowsizes = c(5, 900, 3600),
+        desiredtz = "America/Chicago",
+        do.enmo = TRUE, do.anglez = TRUE,
+        dayborder = 0,
+        strategy = 1, hrs.del.start = 0, hrs.del.end = 0,
+        maxdur = 0, includedaycrit = 0,
+        idloc = 1,
+        dynrange = 8,
+        chunksize = 1,
+        do.cal = TRUE,
+        use.temp = FALSE,
+        spherecrit = 0.3,
+        minloadcrit = 72,
+        printsummary = TRUE,
+        do.imp = TRUE,
+        epochvalues2csv = TRUE,
+        L5M5window = c(0,24),
+        M5L5res = 10,
+        winhr = c(5,10),
+        qlevels = c(960/1440, 1320/1440, 1380/1440, 1410/1440, 1430/1440, 1435/1440, 1438/1440),
+        ilevels = seq(0,600, by = 25),
+        iglevels = c(seq(0,4000, by = 25), 8000),
+        bout.metric = 4,
+        do.visual = TRUE,
+        excludefirstlast = FALSE,
+        includenightcrit = 0,
+        anglethreshold = 5,
+        timethreshold = 5,
+        ignorenonwear = TRUE,
+        acc.metric = "ENMO",
+        do.part3.pdf = TRUE,
+        outliers.only = FALSE,
+        def.noc.sleep = 1,
+        excludefirstlast.part5 = FALSE,
+        maxdur = 0,
+        threshold.lig = c(45),
+        threshold.mod = c(100),
+        threshold.vig = c(430),
+        boutdur.mvpa = c(1,5,10),
+        boutdur.in = c(10,20,30),
+        boutdur.lig = c(1,5,10),
+        boutcriter.mvpa = 0.8,
+        boutcriter.in = 0.9,
+        boutcriter.lig = 0.8,
+        timewindow = c("MM", "WW"),
+        acc.metric = "ENMO",
+        do.report = c(2,4,5),
+        visualreport = TRUE,
+        do.parallel = TRUE
+      ))
+    }
+  }
+}
+
+# Run main if executed as a script
+if (!interactive()) {
+  main()
+}

code/core/acc.R

@@ -1,54 +1,49 @@
-import os
 import subprocess
-import sys
 
 class GG:
+    """
+    Class to execute GGIR processing for matched subject records.
+    """
+
     def __init__(self, matched, intdir, obsdir):
+        """
+        Initialize the GG instance.
+
+        Args:
+            matched (dict): Mapping of subject IDs to their records.
+            intdir (str): Path to the internal directory.
+            obsdir (str): Path to the observational directory.
+        """
         self.matched = matched
-        self.INTDIR = intdir
-        self.OBSDIR = obsdir
+        self.INTDIR = intdir+'/'
+        self.OBSDIR = obsdir+'/'
+        self.DERIVATIVES = "/derivatives/GGIR-3.1.4"  # Defined within the class
 
     def run_gg(self):
-        for subject_id, records in self.matched.items():
-            for record in records:
-                study = record.get('study')
-                file_path = record.get('file_path')
-                session = record.get('run')
-
-                if not file_path or not os.path.exists(file_path):
-                    print(f"Invalid or missing file path for subject {subject_id}")
-                    continue
-
-                # Build the output directory based on study type.
-                if study.lower() == 'obs':
-                    outdir = os.path.join(self.OBSDIR, 'derivatives', 'GGIR-3.1.4',
-                                          f"sub-{subject_id}", "ses-{session}")
-                else:
-                    outdir = os.path.join(self.INTDIR, 'derivatives', 'GGIR-3.1.4',
-                                          f"sub-{subject_id}", "ses-{session}")
-
-                try:
-                    # Create the derivative subdirectory within outdir.
-                    deriv_dir = os.path.join(outdir, 'derivatives')
-
-                    # Construct the Rscript command.
-                    command = (
-                        f"""
-                        Rscript core/basic_accel.R  --input_file {file_path} --output_location {outdir} --verbose
-                        """
-                    )
-
-                    # Run the command in a new subprocess.
-                    print(f"running GGIR for {file_path}")
-                    result = subprocess.run(
-                        command,
-                        check=True,
-                        stdout=sys.stdout,
-                        stderr=sys.stderr
-                    )
-                    # Decode and print the output.
-                    print(f"Command output: {result.stdout.decode().strip()}")
-                except subprocess.CalledProcessError as e:
-                    print(f"Command failed for subject {subject_id} with exit status: {e.returncode}")
-                    print(f"Error output: {e.stderr.decode().strip()}")
-
+        """
+        Run GGIR for both the internal and observational project directories.
+        """
+        for project_dir in [self.INTDIR, self.OBSDIR]:
+            command = f"Rscript core/acc.R --project_dir {project_dir} --deriv_dir {self.DERIVATIVES}"
+
+            try:
+                # Execute the command in a new subprocess
+                print(f"Running GGIR for project directory {project_dir}")
+                process = subprocess.Popen(
+                    command,
+                    shell=True,
+                    stdout=subprocess.PIPE,
+                    stderr=subprocess.STDOUT,
+                    bufsize=1,
+                    universal_newlines=True
+                )
+
+                # Stream output line-by-line
+                for line in process.stdout:
+                    print(line, end='')  # already includes newline
+
+                process.stdout.close()
+                process.wait()
+
+                if process.returncode != 0:
+                    raise subprocess.CalledProcessError(process.returncode, command)