Change the package name from "whims" to "MRIreduce"

.Rproj.user/6A9A55E1/pcs/files-pane.pper

@@ -5,5 +5,5 @@
             "ascending": true
         }
     ],
-    "path": "~/Desktop/WHIMs/R"
+    "path": "~/Desktop/MRIreduce"
 }

.Rproj.user/6A9A55E1/pcs/windowlayoutstate.pper

@@ -1,14 +1,14 @@
 {
     "left": {
-        "splitterpos": 345,
+        "splitterpos": 269,
         "topwindowstate": "NORMAL",
-        "panelheight": 682,
-        "windowheight": 720
+        "panelheight": 663,
+        "windowheight": 701
     },
     "right": {
-        "splitterpos": 373,
+        "splitterpos": 79,
         "topwindowstate": "NORMAL",
-        "panelheight": 682,
-        "windowheight": 720
+        "panelheight": 663,
+        "windowheight": 701
     }
 }

DESCRIPTION

@@ -1,13 +1,13 @@
-Package: whims
+Package: MRIreduce
 Type: Package
-Title: NeuroPartitioner: An R Package for ROI-Based Transformation of Neuroimages into High-Dimensional Data Frames
+Title: MRIreduce: An R Package for ROI-Based Transformation of Neuroimages into High-Dimensional Data Frames
 Version: 0.1.0
 Authors@R: c(
     person("Joshua", "Milstein", email = "[email protected]", role = c("aut")),
     person("Jinyao", "Tian", email = "[email protected]", role = c("aut", "cre"))
   )
 Maintainer: Jinyao Tian <[email protected]>, Joshua Milstein <[email protected]>
-Description: NeuroPartitioner is an advanced R package designed to convert NIfTI format T1/FL neuroimages into structured, high-dimensional 2D data frames, focusing on region of interest (ROI) based processing. This package incorporates a key algorithm called ‘partition’, which offers a fast, flexible framework for agglomerative partitioning based on the Direct-Measure-Reduce approach. This method ensures that each reduced variable maintains a user-specified minimum level of information while being interpretable, as each maps uniquely to one variable in the reduced dataset. The ‘partition’ algorithm, detailed in Millstein et al. (2020), allows for customization in variable selection, measurement of information loss, and data reduction methods. NeuroPartitioner is indispensable for researchers requiring efficient, accurate preparation of neuroimaging data for detailed statistical analysis and machine learning applications, enhancing the interpretability and utility of neuroimaging studies.
+Description: MRIreduce is an advanced R package designed to convert NIfTI format T1/FL neuroimages into structured, high-dimensional 2D data frames, focusing on region of interest (ROI) based processing. This package incorporates a key algorithm called ‘partition’, which offers a fast, flexible framework for agglomerative partitioning based on the Direct-Measure-Reduce approach. This method ensures that each reduced variable maintains a user-specified minimum level of information while being interpretable, as each maps uniquely to one variable in the reduced dataset. The ‘partition’ algorithm, detailed in Millstein et al. (2020), allows for customization in variable selection, measurement of information loss, and data reduction methods. NeuroPartitioner is indispensable for researchers requiring efficient, accurate preparation of neuroimaging data for detailed statistical analysis and machine learning applications, enhancing the interpretability and utility of neuroimaging studies.
 License: MIT + file LICENSE
 Imports:
     Rcpp,
@@ -35,4 +35,4 @@ Suggests:
     knitr,
     rmarkdown
 VignetteBuilder: knitr
-URL: https://jtian123.github.io/WHIMs/
+URL: https://github.com/jtian123/MRIreduce

NAMESPACE

@@ -3,6 +3,7 @@
 export(PartitionPipeline)
 export(eve_Fl)
 export(eve_T1)
+export(map_feature2_loc)
 export(run_partition_pipeline)
 export(suppar)
 import(R6)

R/RcppExports.R

@@ -2,6 +2,6 @@
 # Generator token: 10BE3573-1514-4C36-9D1C-5A225CD40393
 
 partagg <- function(grpl1, corrbin, B) {
-    .Call('_whims_partagg', PACKAGE = 'whims', grpl1, corrbin, B)
+    .Call(`_MRIreduce_partagg`, grpl1, corrbin, B)
 }
 

R/eve_Fl.R

@@ -20,10 +20,12 @@
 #' Volumes are calculated based on the segmented tissues.
 #'
 #' @examples
+#' ## Not run:
 #' eve_Fl("path/to/your/flair/image.nii.gz",
 #'        "path/to/output/",
 #'        "/usr/local/fsl",
 #'        "NIFTI_GZ")
+#' ## End(Not run)
 #'
 #' @importFrom neurobase readnii writenii
 #' @importFrom extrantsr bias_correct fslbet_robust
@@ -114,7 +116,7 @@ eve_Fl <- function(fpath,outpath, fsl_path, fsl_outputtype = "NIFTI_GZ") {
 
   # output array intensities and tissues registered to Eve
   outp <- vector("list", 3)
-  names(outp) <- c("intenisites", "tissues", "brain_volume_cm3")
+  names(outp) <- c("intensities", "tissues", "brain_volume_cm3")
   outp[[1]] <- adat
   outp[[2]] <- adat_fast
   outp[[3]] <- c(vres, bv, total_icv)

R/eve_T1.R

@@ -32,7 +32,9 @@
 #' Each step logs its progress with timestamps, aiding in debugging and optimization of processing times.
 #'
 #' @examples
+#' ## Not run:
 #' eve_T1("path/to/your/image.nii.gz", "path/to/output", "/usr/local/fsl",  "NIFTI_GZ")
+#' ## End(Not run)
 #'
 #' @importFrom neurobase readnii writenii
 #' @importFrom fslr fslreorient2std fsl_biascorrect fslbet fast fslstats
@@ -120,7 +122,7 @@ eve_T1 <- function(fpath, outpath, fsl_path = '/Users/jinyaotian/fsl', fsl_outpu
 
   # output array intensities and tissues registered to Eve
   outp <- vector("list", 3)
-  names(outp) <- c("intenisites", "tissues", "brain_volume_cm3")
+  names(outp) <- c("intensities", "tissues", "brain_volume_cm3")
   outp[[1]] <- adat
   outp[[2]] <- adat_fast
   outp[[3]] <- c(vres,bv, total_icv)

R/funcs.R

@@ -18,8 +18,8 @@ load_required_data <- function(filename, package_name, envir = .GlobalEnv) {
 iproc <- function(tind, nfl, main_dir, outp_volume = TRUE) {
   # Load Eve labels
   # Load necessary data
-  load_required_data("eve_label_array.RData", "whims")
-  load_required_data("eve_label_info_dataframe.RData", "whims")
+  load_required_data("eve_label_array.RData", "MRIreduce")
+  load_required_data("eve_label_info_dataframe.RData", "MRIreduce")
   # Initial checks and setup
   if (!dir.exists(main_dir)) {
     stop("main_dir does not exist: ", main_dir)
@@ -283,7 +283,7 @@ tissue_segment <- function(liste, thresh_vec, tind, tissue_type, main_dir) {
     colnames(count_voxel_matrix) <- par_map[, 'variable']
 
     intensity_df <- as.data.frame(intensity_mean_matrix)
-    intensity_df$fname<- img_name
+    row.names(intensity_df)<- img_name
     volume_df <- as.data.frame(count_voxel_matrix)
     volume_df$fname <- img_name
 
@@ -300,8 +300,12 @@ tissue_segment <- function(liste, thresh_vec, tind, tissue_type, main_dir) {
       merged_df[cols_to_multiply] <- merged_df[cols_to_multiply] * merged_df$unit_voxel
       # Resulting data frame has the product, and you might want to remove the extra unit_voxel column if not needed anymore
       final_df <- merged_df[, !names(merged_df) %in% c("unit_voxel", "brainVolume")]
+      row.names(final_df) = final_df$fname
+      final_df$fname = NULL
       saveRDS(final_df, file = volume_file_path)
     } else {
+      row.names(volume_df) = volume_df$fname
+      volume_df$fname = NULL
       saveRDS(volume_df, file = volume_file_path)
     }
 
@@ -311,15 +315,38 @@ tissue_segment <- function(liste, thresh_vec, tind, tissue_type, main_dir) {
   rslts <- lapply(thresh_vec, process_threshold)
 }
 
+##Test
+# main_dir = '/Users/jinyaotian/Downloads/whims_test'
+# roi = 'inferior_frontal_gyrus_left'
+# dep_list_path = file.path(main_dir, "dep_list", paste0(roi,'.rds'))
+# listes <- readRDS(dep_list_path)
+# cat("tissue segmentation for all sublists based on super partitions.\n")
+# lapply(listes, function(sub_list) {
+#   tissue_segment(liste = sub_list,thresh_vec = 0.8,tind = 5, tissue_type =2, main_dir = main_dir )
+# })
+
 #Combine by tissue type for each threshold and roi
 Cmb_tissue_type <- function(thresh, roi, tissue_type, main_dir){
   modify_data <- function(data, roi_name) {
     colnames(data) <- paste(roi_name, colnames(data), sep = "_")
     return(data)
   }
-  read_rds_file <- function(file_path, intensity = TRUE) {
+  read_rds_file <- function(file_path) {
+    # Extract the file name
+    doc_name <- basename(file_path)
+    # Extract the number from the file name using regex
+    number <- gsub("\\D", "", doc_name)  # Remove non-numeric characters
     # Read the .rds file
     df <- readRDS(file_path)
+    # Update the column names based on whether they contain "reduced_var"
+    colnames(df) <- sapply(colnames(df), function(col_name) {
+      if (grepl("reduced_var", col_name)) {
+        return(paste0("module", number, "_", col_name))
+      } else {
+        return(col_name)
+      }
+    })
+    return(df)  # Return the modified dataframe
   }
 
   if (tissue_type == 3){
@@ -339,12 +366,10 @@ Cmb_tissue_type <- function(thresh, roi, tissue_type, main_dir){
   volume_files_with_prefix <- list.files(path = directory, pattern = paste0("^volume_"), full.names = TRUE)
 
   data_frames_intensity <- lapply(intensity_files_with_prefix, read_rds_file)
-  data_frames_volume <- lapply(volume_files_with_prefix, function(file) read_rds_file(file, intensity = FALSE))
+  data_frames_volume <- lapply(volume_files_with_prefix, read_rds_file)
 
   processed_data_frames_intensity <- lapply(data_frames_intensity, function(ls_) {
     df <- ls_
-    row.names(df) <- df$fname
-    df$fname <- NULL
     modify_data(df, roi_name = roi)
   })
 
@@ -366,6 +391,9 @@ Cmb_tissue_type <- function(thresh, roi, tissue_type, main_dir){
   saveRDS(combined_data_frame_volume, file = file.path(save.dir,"volume.rds"))
 }
 
+##Test
+#Cmb_tissue_type(thresh = 0.8, roi = roi, tissue_type = 2, main_dir = main_dir)
+
 #Process independent variables from Super-Partition
 process_indep_variables <- function(indep_list, tissue_type, tind, roi, main_dir) {
   if (tissue_type == 3){
@@ -463,6 +491,9 @@ Cmb_indep_with_dep <- function(tissue_type, roi, thresh, main_dir){
   }
 }
 
+##Test
+#Cmb_indep_with_dep(tissue_type = 2, roi = "inferior_frontal_gyrus_left", thresh = 0.8, main_dir = '/Users/jinyaotian/Downloads/whims_test')
+
 #Combine by ROI
 concat_reduced_var <- function(roi, thresh, main_dir){
   #Structure of dir: /roi/thresh

R/map_feature2_loc.R

@@ -0,0 +1,68 @@
+#' Map Reduced Feature Back to Brain Image Voxel Locations
+#'
+#' This function maps a given feature name to voxel locations by identifying the pattern in the feature name.
+#' If the feature name contains "reduced_var", it follows the format "Roi_module_reduced_var" and this indicates that it is a reduced feature. Otherwise, it follows
+#' the format "Roi_Vnumber", which is not a reduced feature by Partition. Based on the pattern, the function extracts the appropriate region of interest (ROI) and
+#' retrieves voxel locations from intensity data.
+#'
+#' @param feature_name String. A feature name.
+#' @param threshold a numeric between 0 and 1. The Partition threshold value applied to the data.
+#' @param main_dir String. The main directory containing the data files.
+#'
+#' @return A data frame containing the voxel locations (x,y,z Coordinates) corresponding to the extracted mapping column.
+#'
+#' @details
+#' Each voxel in a brain image corresponds to a specific feature, establishing a one-to-one mapping between a voxel and a feature. This relationship allows us
+#' to localize particular brain features to specific brain areas at the voxel level, enabling visualization of these features. However, after applying Super-Partition and Partition techniques,
+#' multiple brain features are aggregated into a single reduced feature as part of a data reduction process. The goal of the following function is to relate the reduced feature back to its component features,
+#' and subsequently identify the brain image voxels to which those component features are mapped.
+#' This function performs the following steps:
+#'
+#' 1. **Check Feature Name Format:** It first checks if the `feature_name` contains "reduced_var". If it does, the function assumes the format "Roi_module_reduced_var", otherwise it assumes the format "Roi_Vnumber".
+#'
+#' 2. **Extract ROI and Mapping Information:**
+#'    - For the "Roi_module_reduced_var" format, the ROI and module number are extracted, and the corresponding partition file is read to retrieve the mapping vector.
+#'    - For the "Roi_Vnumber" format, the `V` number is extracted from the feature name and used as the mapping vector.
+#'
+#' 3. **Extract Voxel Locations:** The mapping vector is then used to extract voxel locations from the intensity data.
+#'
+#' @examples
+#' \dontrun{
+#' loc_df <- map_feature2_loc(feature_name = "inferior_frontal_gyrus_left_module4_reduced_var_13", threshold = 0.8, main_dir = "/path/to/data")
+#' }
+#'
+#' @export
+map_feature2_loc <- function(feature_name, threshold,main_dir) {
+  if (grepl("reduced_var", feature_name)) {
+    # Extract the ROI for the format "Roi_module_reduced_var"
+    roi <- sub("_module.*", "", feature_name)  # Remove everything after and including "_module"
+    module = sub(".*module(\\d+).*", "\\1", feature_name)
+    variable_name = sub(".*(reduced_var_\\d+).*", "\\1", feature_name)
+    file_path = file.path(main_dir, "partition", roi, threshold, "map", paste0("Par_intensities_map_", module, ".rds"))
+    df = readRDS(file_path)
+    vec <- df[df$variable == variable_name, "mapping"][[1]]
+  } else {
+    # Extract the ROI for the format "Roi_Vnumber"
+    roi <- sub("_V\\d+.*", "", feature_name)  # Remove everything after and including "_V<number>"
+    vec <- sub(".*(V\\d+).*", "\\1", feature_name)
+  }
+
+  # Extract voxel locations
+  extract_mapping_voxel <- function(tind, roi, Cols) {
+    file_name <- paste0("intensities_", roi, "_", tind,".rds")
+    df <- readRDS(file.path(main_dir, "intensities", file_name))
+    locs <- df[1:3, Cols, drop = FALSE]
+    return(locs)
+  }
+  load_required_data("eve_label_info_dataframe.RData", "whims")
+  tind = lab_df[lab_df$text_label == roi, "integer_label"]
+  # Extract the voxel locations
+  loc_df <- extract_mapping_voxel(tind, roi, vec)
+  return(loc_df)
+}
+
+##Test
+# feature_name = "inferior_frontal_gyrus_left_module11_reduced_var_5"
+# threshold = 0.8
+# main_dir = '/Users/jinyaotian/Downloads/whims_test'
+# loc_df = map_feature2_loc(feature_name = feature_name, threshold = threshold, main_dir = main_dir)

README.Rmd

@@ -1,5 +1,5 @@
 ---
-title: "NeuroPartitioner"
+title: "MRIreduce"
 output: github_document
 ---
 
@@ -13,7 +13,7 @@ knitr::opts_chunk$set(echo = TRUE)
 
 ## Introduction
 
-NeuroPartitioner is an advanced R package designed to convert NIfTI format T1/FL neuroimages into structured, high-dimensional 2D data frames, focusing on region of interest (ROI) based processing. This package incorporates a key algorithm called ‘partition’, which offers a fast, flexible framework for agglomerative partitioning based on the Direct-Measure-Reduce approach. This method ensures that each reduced variable maintains a user-specified minimum level of information while being interpretable, as each maps uniquely to one variable in the reduced dataset. The ‘partition’ algorithm, detailed in Millstein et al. (2020), allows for customization in variable selection, measurement of information loss, and data reduction methods. NeuroPartitioner is indispensable for researchers requiring efficient, accurate preparation of neuroimaging data for detailed statistical analysis and machine learning applications, enhancing the interpretability and utility of neuroimaging studies.
+MRIreduce is an advanced R package designed to convert NIfTI format T1/FL neuroimages into structured, high-dimensional 2D data frames, focusing on region of interest (ROI) based processing. This package incorporates a key algorithm called ‘partition’, which offers a fast, flexible framework for agglomerative partitioning based on the Direct-Measure-Reduce approach. This method ensures that each reduced variable maintains a user-specified minimum level of information while being interpretable, as each maps uniquely to one variable in the reduced dataset. The ‘partition’ algorithm, detailed in Millstein et al. (2020), allows for customization in variable selection, measurement of information loss, and data reduction methods. MRIreduce is indispensable for researchers requiring efficient, accurate preparation of neuroimaging data for detailed statistical analysis and machine learning applications, enhancing the interpretability and utility of neuroimaging studies.
 
 
 ## Installation Instructions

README.md

@@ -1,4 +1,4 @@
-NeuroPartitioner
+MRIreduce
 ================
 
 <!-- badges: start -->
@@ -9,17 +9,17 @@ IMAGE](https://raw.githubusercontent.com/USCbiostats/badges/master/tommy-image-b
 
 ## Introduction
 
-NeuroPartitioner is an advanced R package designed to convert NIfTI
-format T1/FL neuroimages into structured, high-dimensional 2D data
-frames, focusing on region of interest (ROI) based processing. This
-package incorporates a key algorithm called ‘partition’, which offers a
-fast, flexible framework for agglomerative partitioning based on the
+MRIreduce is an advanced R package designed to convert NIfTI format
+T1/FL neuroimages into structured, high-dimensional 2D data frames,
+focusing on region of interest (ROI) based processing. This package
+incorporates a key algorithm called ‘partition’, which offers a fast,
+flexible framework for agglomerative partitioning based on the
 Direct-Measure-Reduce approach. This method ensures that each reduced
 variable maintains a user-specified minimum level of information while
 being interpretable, as each maps uniquely to one variable in the
 reduced dataset. The ‘partition’ algorithm, detailed in Millstein et
 al. (2020), allows for customization in variable selection, measurement
-of information loss, and data reduction methods. NeuroPartitioner is
+of information loss, and data reduction methods. MRIreduce is
 indispensable for researchers requiring efficient, accurate preparation
 of neuroimaging data for detailed statistical analysis and machine
 learning applications, enhancing the interpretability and utility of

_pkgdown.yml

@@ -1,18 +1,18 @@
-url: https://jtian123.github.io/WHIMs/
+url: https://jtian123.github.io/MRIreduce/
 template:
   bootstrap: 5
   bootswatch: cosmo  # Using the Flatly theme from Bootswatch
 
 navbar:
-  title: NeuroPartitioner
+  title: MRIreduce
   left:
     - text: Home
       href: index.html
     - text: Get Started
-      href: articles/NeuroPartitioner.html  # Link directly to your vignette
+      href: articles/MRIreduce.html  # Link directly to your vignette
     - text: Reference
       href: reference/index.html
   right:
     - text: View on source code
-      href: https://github.com/jtian123/WHIMs
+      href: https://github.com/jtian123/MRIreduce
       icon: fab fa-github