From 281eba2b00588f05e928787c7ef969f1ff2a7bee Mon Sep 17 00:00:00 2001
From: delfarahalireza <delfarah@calicolabs.com>
Date: Mon, 6 Nov 2023 11:51:57 -0800
Subject: [PATCH 1/7] issue-21 updated floor_peaks() function

---
 R/mutate_mzroll_list.R | 105 ++++++++++++++++++++++++++++-------------
 1 file changed, 73 insertions(+), 32 deletions(-)

diff --git a/R/mutate_mzroll_list.R b/R/mutate_mzroll_list.R
index 0e1bb02..51a87a7 100644
--- a/R/mutate_mzroll_list.R
+++ b/R/mutate_mzroll_list.R
@@ -4,55 +4,96 @@
 #'   undetected peaks.
 #'
 #' @inheritParams test_mzroll_list
-#' @param log2_floor_value minimum value to set for low abundance or
-#'   missing peaks
+#' @param floor_value minimum value to set for low abundance or missing peaks
 #' @param floor_var measurement variable to floor to \code{log2_floor_value}.
+#' @param mzrolldb_file_path (optional) path to mzrollDB file, only used for QQQ datasets.
+#' @param data_type parameter to differentiate data types.
 #'
 #' @return \code{\link{process_mzroll}}
 #'
 #' @examples
 #' floored_peaks <- floor_peaks(nplug_mzroll_augmented, 12)
+#' floored_peaks <- floor_peaks(mzroll_list, 100, "log2_abundance", "QQQ")
+
 #' @export
 floor_peaks <- function(mzroll_list,
-                        log2_floor_value = 12,
-                        floor_var = "log2_abundance") {
-  test_mzroll_list(mzroll_list)
-
+                         floor_value = 12,
+                         floor_var = "log2_abundance",
+                         mzrolldb_file_path = NULL,
+                         data_type = "QE") {
+  test_mzroll_list_local(mzroll_list)
+  
   valid_floor_var <- setdiff(
     mzroll_list$design$measurements$variable,
     c(mzroll_list$design$feature_pk, mzroll_list$design$sample_pk)
   )
-
+  
   checkmate::assertChoice(floor_var, valid_floor_var)
   checkmate::assertNumeric(mzroll_list$measurements[[floor_var]])
-  checkmate::assertNumber(log2_floor_value)
-
-  # summarize peaks associated with each peakgroup
-
-  missing_peaks <- tidyr::expand_grid(
-    groupId = mzroll_list$features$groupId,
-    sampleId = mzroll_list$samples$sampleId
-  ) %>%
-    dplyr::anti_join(
+  checkmate::assertNumber(floor_value)
+  
+  if (data_type == "QQQ") {
+    ## find missing peaks
+    missing_peaks <- tidyr::expand_grid(
+      groupId = mzroll_list$features$groupId,
+      sampleId = mzroll_list$samples$sampleId) %>%
+      dplyr::anti_join (
+        mzroll_list$measurements,
+        by = c("groupId", "sampleId")
+      )
+    
+    ## find groupBackground for missing peaks
+    ## assign 100 if the groupBackground is zero
+    group_background <- PDB_import(mzrolldb_file_path) %>%
+      group_by(groupId) %>%
+      distinct(groupId, .keep_all = TRUE) %>%
+      mutate(groupBackground = dplyr::case_when(
+        groupBackground < floor_value ~ floor_value,
+        TRUE ~ groupBackground)) %>%
+      mutate(log2_abundance = log2(groupBackground)) %>%
+      mutate(groupId = factor(groupId)) %>%
+      select(groupId, log2_abundance)
+    
+    ## impute missing peaks
+    missing_peaks_imputed <- left_join(
+      missing_peaks, 
+      group_background, 
+      by = c("groupId")) %>%
+      rowwise() %>%
+      mutate(!!rlang::sym(imputation_var) := rnorm(1, mean = !!rlang::sym(imputation_var)+1, sd = 0.1))
+    
+    ## merge measured peaks with imputed peaks
+    completed_peaks <- dplyr::bind_rows(
       mzroll_list$measurements,
-      by = c("groupId", "sampleId")
+      missing_peaks_imputed
+    )
+  } 
+  else {
+    missing_peaks <- tidyr::expand_grid(
+      groupId = mzroll_list$features$groupId,
+      sampleId = mzroll_list$samples$sampleId
+    ) %>%
+      dplyr::anti_join(
+        mzroll_list$measurements,
+        by = c("groupId", "sampleId")
       ) %>%
-    tibble::as_tibble() %>%
-    dplyr::mutate(!!rlang::sym(floor_var) := log2_floor_value)
-
-  # combine detected peaks with peaks that were missing for some samples
-  completed_peaks <- dplyr::bind_rows(
-    mzroll_list$measurements %>%
-      dplyr::mutate(!!rlang::sym(floor_var) := dplyr::case_when(
-        is.na(!!rlang::sym(floor_var)) ~ log2_floor_value,
-        !!rlang::sym(floor_var) < log2_floor_value ~ log2_floor_value,
-        !!rlang::sym(floor_var) >= log2_floor_value ~ !!rlang::sym(floor_var)
-      )),
-    missing_peaks
-  )
-
+      tibble::as_tibble() %>%
+      dplyr::mutate(!!rlang::sym(floor_var) := floor_value)
+    
+    # combine detected peaks with peaks that were missing for some samples
+    completed_peaks <- dplyr::bind_rows(
+      mzroll_list$measurements %>%
+        dplyr::mutate(!!rlang::sym(floor_var) := dplyr::case_when(
+          is.na(!!rlang::sym(floor_var)) ~ floor_value,
+          !!rlang::sym(floor_var) < floor_value ~ floor_value,
+          !!rlang::sym(floor_var) >= floor_value ~ !!rlang::sym(floor_var)
+        )),
+      missing_peaks
+    )
+  }
+  
   mzroll_list$measurements <- completed_peaks
-
+  
   return(mzroll_list)
 }
 

From f88b1568b6feeaa7804c27689ba0272453135d4e Mon Sep 17 00:00:00 2001
From: delfarahalireza <delfarah@calicolabs.com>
Date: Thu, 16 Nov 2023 17:35:55 -0800
Subject: [PATCH 2/7] functions to  fill in missing peaks

---
 NAMESPACE                    |   2 +
 R/mutate_mzroll_list.R       | 180 +++++++++++++++++++++--------------
 man/fill_in_missing_peaks.Rd |  33 +++++++
 man/impute_missing_peaks.Rd  |  32 +++++++
 4 files changed, 178 insertions(+), 69 deletions(-)
 create mode 100644 man/fill_in_missing_peaks.Rd
 create mode 100644 man/impute_missing_peaks.Rd

diff --git a/NAMESPACE b/NAMESPACE
index f66f484..7babc77 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -3,9 +3,11 @@
 export(collapse_injections)
 export(collapse_metabolites)
 export(diffex_mzroll)
+export(fill_in_missing_peaks)
 export(filter_groupIds)
 export(find_pathway_enrichments)
 export(floor_peaks)
+export(impute_missing_peaks)
 export(is_has_label)
 export(lipid_components)
 export(merge_compounds_tbl)
diff --git a/R/mutate_mzroll_list.R b/R/mutate_mzroll_list.R
index 51a87a7..c755655 100644
--- a/R/mutate_mzroll_list.R
+++ b/R/mutate_mzroll_list.R
@@ -4,24 +4,19 @@
 #'   undetected peaks.
 #'
 #' @inheritParams test_mzroll_list
-#' @param floor_value minimum value to set for low abundance or missing peaks
+#' @param log2_floor_value minimum value to set for low abundance or
+#'   missing peaks
 #' @param floor_var measurement variable to floor to \code{log2_floor_value}.
-#' @param mzrolldb_file_path (optional) path to mzrollDB file, only used for QQQ datasets.
-#' @param data_type parameter to differentiate data types.
 #'
 #' @return \code{\link{process_mzroll}}
 #'
 #' @examples
 #' floored_peaks <- floor_peaks(nplug_mzroll_augmented, 12)
-#' floored_peaks <- floor_peaks(mzroll_list, 100, "log2_abundance", "QQQ")
-
 #' @export
 floor_peaks <- function(mzroll_list,
-                         floor_value = 12,
-                         floor_var = "log2_abundance",
-                         mzrolldb_file_path = NULL,
-                         data_type = "QE") {
-  test_mzroll_list_local(mzroll_list)
+                        log2_floor_value = 12,
+                        floor_var = "log2_abundance") {
+  test_mzroll_list(mzroll_list)
   
   valid_floor_var <- setdiff(
     mzroll_list$design$measurements$variable,
@@ -30,73 +25,120 @@ floor_peaks <- function(mzroll_list,
   
   checkmate::assertChoice(floor_var, valid_floor_var)
   checkmate::assertNumeric(mzroll_list$measurements[[floor_var]])
-  checkmate::assertNumber(floor_value)
-  
-  if (data_type == "QQQ") {
-    ## find missing peaks
-    missing_peaks <- tidyr::expand_grid(
-      groupId = mzroll_list$features$groupId,
-      sampleId = mzroll_list$samples$sampleId) %>%
-      dplyr::anti_join (
-        mzroll_list$measurements,
-        by = c("groupId", "sampleId")
-      )
-    
-    ## find groupBackground for missing peaks
-    ## assign 100 if the groupBackground is zero
-    group_background <- PDB_import(mzrolldb_file_path) %>%
-      group_by(groupId) %>%
-      distinct(groupId, .keep_all = TRUE) %>%
-      mutate(groupBackground = dplyr::case_when(
-        groupBackground < floor_value ~ floor_value,
-        TRUE ~ groupBackground)) %>%
-      mutate(log2_abundance = log2(groupBackground)) %>%
-      mutate(groupId = factor(groupId)) %>%
-      select(groupId, log2_abundance)
-    
-    ## impute missing peaks
-    missing_peaks_imputed <- left_join(
-      missing_peaks, 
-      group_background, 
-      by = c("groupId")) %>%
-      rowwise() %>%
-      mutate(!!rlang::sym(imputation_var) := rnorm(1, mean = !!rlang::sym(imputation_var)+1, sd = 0.1))
-    
-    ## merge measured peaks with imputed peaks
-    completed_peaks <- dplyr::bind_rows(
+  checkmate::assertNumber(log2_floor_value)
+  
+  # summarize peaks associated with each peakgroup
+  
+  missing_peaks <- tidyr::expand_grid(
+    groupId = mzroll_list$features$groupId,
+    sampleId = mzroll_list$samples$sampleId
+  ) %>%
+    dplyr::anti_join(
       mzroll_list$measurements,
-      missing_peaks_imputed
-    )
-  } 
-  else {
-    missing_peaks <- tidyr::expand_grid(
-      groupId = mzroll_list$features$groupId,
-      sampleId = mzroll_list$samples$sampleId
+      by = c("groupId", "sampleId")
     ) %>%
-      dplyr::anti_join(
-        mzroll_list$measurements,
-        by = c("groupId", "sampleId")
-      ) %>%
-      tibble::as_tibble() %>%
-      dplyr::mutate(!!rlang::sym(floor_var) := floor_value)
-    
-    # combine detected peaks with peaks that were missing for some samples
-    completed_peaks <- dplyr::bind_rows(
-      mzroll_list$measurements %>%
-        dplyr::mutate(!!rlang::sym(floor_var) := dplyr::case_when(
-          is.na(!!rlang::sym(floor_var)) ~ floor_value,
-          !!rlang::sym(floor_var) < floor_value ~ floor_value,
-          !!rlang::sym(floor_var) >= floor_value ~ !!rlang::sym(floor_var)
-        )),
-      missing_peaks
-    )
-  }
+    tibble::as_tibble() %>%
+    dplyr::mutate(!!rlang::sym(floor_var) := log2_floor_value)
+  
+  # combine detected peaks with peaks that were missing for some samples
+  completed_peaks <- dplyr::bind_rows(
+    mzroll_list$measurements %>%
+      dplyr::mutate(!!rlang::sym(floor_var) := dplyr::case_when(
+        is.na(!!rlang::sym(floor_var)) ~ log2_floor_value,
+        !!rlang::sym(floor_var) < log2_floor_value ~ log2_floor_value,
+        !!rlang::sym(floor_var) >= log2_floor_value ~ !!rlang::sym(floor_var)
+      )),
+    missing_peaks
+  )
   
   mzroll_list$measurements <- completed_peaks
   
   return(mzroll_list)
 }
 
+#' Impute missing peaks with provided feature-level imputation values
+#'
+#'@param mzroll_list: data in triple omic structure
+#'@param lod_values: a tibble that maps groupId to log2 feature-level imputation values
+#'@param quant_var: column to use for peak values
+#'@param imputation_sd: standard deviation of Gaussian distribution to use for missing peak imputation
+#'
+#'@return triple omic data with imputed missing peaks
+#'
+#' @examples
+#' mzroll_list_imputed <- impute_missing_peaks(mzroll_list, lod_values, "log2_abundance", 0.15)
+#'
+#'@export
+impute_missing_peaks <- function(mzroll_list,
+                                 lod_values,
+                                 quant_var = "log2_abundance",
+                                 imputation_sd = 0.15) {
+  
+  test_mzroll_list(mzroll_list)
+  
+  valid_quant_var <- setdiff(
+    mzroll_list$design$measurements$variable,
+    c(mzroll_list$design$feature_pk, mzroll_list$design$sample_pk)
+  )
+  
+  checkmate::assertChoice(quant_var, valid_quant_var)
+  checkmate::assertNumeric(mzroll_list$measurements[[quant_var]])
+  
+  ## find missing peaks
+  missing_peaks <- tidyr::expand_grid(
+    groupId = mzroll_list$features$groupId,
+    sampleId = mzroll_list$samples$sampleId) %>%
+    dplyr::anti_join (
+      mzroll_list$measurements,
+      by = c("groupId", "sampleId")
+    )
+  
+  ## impute missing peaks
+  missing_peaks_imputed <- left_join(
+    missing_peaks, 
+    group_background, 
+    by = c("groupId")) %>%
+    rowwise() %>%
+    mutate(!!rlang::sym(quant_var) := rnorm(1, mean = !!rlang::sym(quant_var)+1, sd = imputation_sd))
+  
+  ## merge measured peaks with imputed peaks
+  completed_peaks <- dplyr::bind_rows(
+    mzroll_list$measurements,
+    missing_peaks_imputed
+  )
+  
+  mzroll_list$measurements <- completed_peaks
+  return(mzroll_list)
+}
+
+#' Fill in missing peaks
+#'
+#'@param mzroll_list: data in triple omic structure
+#'@param fill_values: either a numeric constant or a tibble that maps groupId to log2 feature-level imputation values
+#'@param quant_var: column to use for peak values
+#'@param imputation_sd: standard deviation of Gaussian distribution to use for missing peak imputation
+#'
+#'@return triple omic data with imputed missing peaks
+#'
+#' @examples
+#' mzroll_list_imputed <- fill_in_missing_peaks(mzroll_list, lod_values, "log2_abundance", 0.15)
+#' mzroll_list_imputed <- fill_in_missing_peaks(mzroll_list, 12, "log2_abundance")
+#'
+#'@export
+fill_in_missing_peaks <- function(mzroll_list,
+                                  fill_values,
+                                  quant_var = "log2_abundance",
+                                  imputation_sd = 0.15) {
+  if (is.data.frame(fill_values)) {
+    stopifnot(colnames(fill_values) %in% c("groupId", rlang::sym(quant_var)))
+    output <- temp_impute(mzroll_list, lod_values, quant_var, imputation_sd)
+  } else if (is.numeric(fill_values)) {
+    output <- floor_peaks(mzroll_list, fill_values, quant_var)
+  } else {
+    stop("fill_values should either be a single numeric value or a tibble that maps groupId to feature-level imputation values")
+  }
+  return(output)
+}
 
 #' Normalize Peaks
 #'
diff --git a/man/fill_in_missing_peaks.Rd b/man/fill_in_missing_peaks.Rd
new file mode 100644
index 0000000..4acc41f
--- /dev/null
+++ b/man/fill_in_missing_peaks.Rd
@@ -0,0 +1,33 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/mutate_mzroll_list.R
+\name{fill_in_missing_peaks}
+\alias{fill_in_missing_peaks}
+\title{Fill in missing peaks}
+\usage{
+fill_in_missing_peaks(
+  mzroll_list,
+  fill_values,
+  quant_var = "log2_abundance",
+  imputation_sd = 0.15
+)
+}
+\arguments{
+\item{mzroll_list:}{data in triple omic structure}
+
+\item{fill_values:}{either a numeric constant or a tibble that maps groupId to log2 feature-level imputation values}
+
+\item{quant_var:}{column to use for peak values}
+
+\item{imputation_sd:}{standard deviation of Gaussian distribution to use for missing peak imputation}
+}
+\value{
+triple omic data with imputed missing peaks
+}
+\description{
+Fill in missing peaks
+}
+\examples{
+mzroll_list_imputed <- fill_in_missing_peaks(mzroll_list, lod_values, "log2_abundance", 0.15)
+mzroll_list_imputed <- fill_in_missing_peaks(mzroll_list, 12, "log2_abundance")
+
+}
diff --git a/man/impute_missing_peaks.Rd b/man/impute_missing_peaks.Rd
new file mode 100644
index 0000000..c3e9c3a
--- /dev/null
+++ b/man/impute_missing_peaks.Rd
@@ -0,0 +1,32 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/mutate_mzroll_list.R
+\name{impute_missing_peaks}
+\alias{impute_missing_peaks}
+\title{Impute missing peaks with provided feature-level imputation values}
+\usage{
+impute_missing_peaks(
+  mzroll_list,
+  lod_values,
+  quant_var = "log2_abundance",
+  imputation_sd = 0.15
+)
+}
+\arguments{
+\item{mzroll_list:}{data in triple omic structure}
+
+\item{lod_values:}{a tibble that maps groupId to log2 feature-level imputation values}
+
+\item{quant_var:}{column to use for peak values}
+
+\item{imputation_sd:}{standard deviation of Gaussian distribution to use for missing peak imputation}
+}
+\value{
+triple omic data with imputed missing peaks
+}
+\description{
+Impute missing peaks with provided feature-level imputation values
+}
+\examples{
+mzroll_list_imputed <- impute_missing_peaks(mzroll_list, lod_values, "log2_abundance", 0.15)
+
+}

From 6e8d2b679ec1418abe497decd59bded8ee3589ae Mon Sep 17 00:00:00 2001
From: delfarahalireza <delfarah@calicolabs.com>
Date: Thu, 16 Nov 2023 17:41:54 -0800
Subject: [PATCH 3/7] fixing an error in the new function

---
 R/mutate_mzroll_list.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/R/mutate_mzroll_list.R b/R/mutate_mzroll_list.R
index c755655..7c21cca 100644
--- a/R/mutate_mzroll_list.R
+++ b/R/mutate_mzroll_list.R
@@ -131,7 +131,7 @@ fill_in_missing_peaks <- function(mzroll_list,
                                   imputation_sd = 0.15) {
   if (is.data.frame(fill_values)) {
     stopifnot(colnames(fill_values) %in% c("groupId", rlang::sym(quant_var)))
-    output <- temp_impute(mzroll_list, lod_values, quant_var, imputation_sd)
+    output <- impute_missing_peaks(mzroll_list, lod_values, quant_var, imputation_sd)
   } else if (is.numeric(fill_values)) {
     output <- floor_peaks(mzroll_list, fill_values, quant_var)
   } else {

From 3f74b800eb8124b4d682001111b908b8a1bd5c82 Mon Sep 17 00:00:00 2001
From: delfarahalireza <delfarah@calicolabs.com>
Date: Thu, 16 Nov 2023 21:55:16 -0800
Subject: [PATCH 4/7] replacing groupBackground with lod_values

---
 R/mutate_mzroll_list.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/R/mutate_mzroll_list.R b/R/mutate_mzroll_list.R
index 7c21cca..ddd4b8b 100644
--- a/R/mutate_mzroll_list.R
+++ b/R/mutate_mzroll_list.R
@@ -96,7 +96,7 @@ impute_missing_peaks <- function(mzroll_list,
   ## impute missing peaks
   missing_peaks_imputed <- left_join(
     missing_peaks, 
-    group_background, 
+    lod_values, 
     by = c("groupId")) %>%
     rowwise() %>%
     mutate(!!rlang::sym(quant_var) := rnorm(1, mean = !!rlang::sym(quant_var)+1, sd = imputation_sd))

From 2ad8e99b1f30565f2b16e70e0063b962ef0e75c3 Mon Sep 17 00:00:00 2001
From: delfarahalireza <delfarah@calicolabs.com>
Date: Fri, 17 Nov 2023 10:16:59 -0800
Subject: [PATCH 5/7] adding description to new function

---
 R/mutate_mzroll_list.R       | 4 ++++
 man/fill_in_missing_peaks.Rd | 3 ++-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/R/mutate_mzroll_list.R b/R/mutate_mzroll_list.R
index ddd4b8b..6dea3eb 100644
--- a/R/mutate_mzroll_list.R
+++ b/R/mutate_mzroll_list.R
@@ -113,6 +113,10 @@ impute_missing_peaks <- function(mzroll_list,
 
 #' Fill in missing peaks
 #'
+#'@description
+#'If \code{fill_values} is a data frame, this function calls \code{impute_missing_peaks()}.
+#'If it is a numeric vector, 'this function calls \code{floor_peaks()}. Other types are not currently supported
+#'
 #'@param mzroll_list: data in triple omic structure
 #'@param fill_values: either a numeric constant or a tibble that maps groupId to log2 feature-level imputation values
 #'@param quant_var: column to use for peak values
diff --git a/man/fill_in_missing_peaks.Rd b/man/fill_in_missing_peaks.Rd
index 4acc41f..dee2c94 100644
--- a/man/fill_in_missing_peaks.Rd
+++ b/man/fill_in_missing_peaks.Rd
@@ -24,7 +24,8 @@ fill_in_missing_peaks(
 triple omic data with imputed missing peaks
 }
 \description{
-Fill in missing peaks
+If \code{fill_values} is a data frame, this function calls \code{impute_missing_peaks()}.
+If it is a numeric vector, 'this function calls \code{floor_peaks()}. Other types are not currently supported
 }
 \examples{
 mzroll_list_imputed <- fill_in_missing_peaks(mzroll_list, lod_values, "log2_abundance", 0.15)

From abe5470cb0246a9b65962629f7b1012e346f34f4 Mon Sep 17 00:00:00 2001
From: delfarahalireza <delfarah@calicolabs.com>
Date: Tue, 21 Nov 2023 15:28:26 -0800
Subject: [PATCH 6/7] issue-21 adding validations for new functions

---
 R/differential_expression.R  | 29 +++++++++++++++------------
 R/mutate_mzroll_list.R       | 39 +++++++++++++++++++++++++++++++-----
 man/fill_in_missing_peaks.Rd |  9 +++++++++
 man/impute_missing_peaks.Rd  |  9 +++++++++
 man/plot_volcano.Rd          | 16 ++++++++-------
 vignettes/NPLUG.Rmd          |  2 +-
 6 files changed, 78 insertions(+), 26 deletions(-)

diff --git a/R/differential_expression.R b/R/differential_expression.R
index 4e5f871..97ae99f 100644
--- a/R/differential_expression.R
+++ b/R/differential_expression.R
@@ -279,15 +279,17 @@ diffex_fdr <- function(term_data) {
 #' @returns a grob
 #'
 #' @examples
+#' library(dplyr)
 #' regression_significance <- diffex_mzroll(
-#'   nplug_mzroll_normalized,
-#'   "normalized_log2_abundance",
-#'   "limitation + limitation:DR + 0",
-#'   FDR_cutoff = 0.01,
-#'   feature_labels = c("UDP", "Lactate", "Serine")
-#' )
-#'
-#' plot_volcano(regression_significance, 10, 0.1)
+#' nplug_mzroll_normalized,
+#' "normalized_log2_abundance",
+#' "limitation + limitation:DR + 0") %>%
+#' dplyr::left_join(
+#' nplug_mzroll_normalized$features %>% select(groupId, compoundName),
+#' by = "groupId")
+#' 
+#' plot_volcano(regression_significance, 10, 0.1, c("Ribose-P", "acetyl-CoA", "ATP"))
+#' 
 #' @export
 plot_volcano <- function(
   regression_significance,
@@ -316,11 +318,12 @@ plot_volcano <- function(
       is_discovery = qvalue < FDR_cutoff
     ) %>%
     ggplot(aes_string(x = effect_var)) +
-    {if ("compoundName" %in% colnames(regression_significance))
-    {geom_point(aes(y = p.value.trans, color = is_discovery, name = compoundName))} 
-      else {geom_point(aes(y = p.value.trans, color = is_discovery))} 
-    } +
-    geom_text(aes(label = ifelse(compoundName %in% feature_labels, compoundName, ""), y = p.value.trans, vjust = -0.75)) +
+    {if ("compoundName" %in% colnames(regression_significance)) {
+      geom_point(aes(y = p.value.trans, color = is_discovery, name = compoundName)) +
+        geom_text(aes(label = ifelse(compoundName %in% feature_labels, compoundName, ""), y = p.value.trans, vjust = -0.75))
+      } 
+      else {geom_point(aes(y = p.value.trans, color = is_discovery))}
+      } +
     facet_wrap(~term, scales = "free_x") +
     scale_x_continuous("Effect size") +
     scale_y_continuous(expression(-log[10] ~ "pvalue")) +
diff --git a/R/mutate_mzroll_list.R b/R/mutate_mzroll_list.R
index 6dea3eb..fe71e5a 100644
--- a/R/mutate_mzroll_list.R
+++ b/R/mutate_mzroll_list.R
@@ -66,6 +66,15 @@ floor_peaks <- function(mzroll_list,
 #'@return triple omic data with imputed missing peaks
 #'
 #' @examples
+#' library(dplyr)
+#' 
+#' lod_values <- nplug_mzroll_augmented[["measurements"]] %>%
+#' dplyr::select(groupId, log2_abundance) %>%
+#' dplyr::distinct(groupId, .keep_all = TRUE)
+#' 
+#' mzroll_list <- nplug_mzroll_augmented
+#' mzroll_list$measurements <- mzroll_list$measurements %>% filter(groupId != 2 & sampleId != 1)
+#' 
 #' mzroll_list_imputed <- impute_missing_peaks(mzroll_list, lod_values, "log2_abundance", 0.15)
 #'
 #'@export
@@ -76,6 +85,18 @@ impute_missing_peaks <- function(mzroll_list,
   
   test_mzroll_list(mzroll_list)
   
+  stopifnot(colnames(lod_values) %in% c("groupId", rlang::sym(quant_var)))
+  
+  if (nrow(lod_values) > nrow(lod_values %>% dplyr::distinct(groupId, .keep_all = TRUE))) {
+    stop("only one value per feature must be provided to impute missing peaks")
+  }
+  
+  features <- mzroll_list$features %>% dplyr::select(groupId)
+  
+  if(!all(features$groupId %in% lod_values$groupId) | nrow(features) != nrow(lod_values)) {
+    stop("groupId values of lod_value table and feature table of triple omic data must match")
+  }
+  
   valid_quant_var <- setdiff(
     mzroll_list$design$measurements$variable,
     c(mzroll_list$design$feature_pk, mzroll_list$design$sample_pk)
@@ -94,12 +115,12 @@ impute_missing_peaks <- function(mzroll_list,
     )
   
   ## impute missing peaks
-  missing_peaks_imputed <- left_join(
+  missing_peaks_imputed <- dplyr::left_join(
     missing_peaks, 
     lod_values, 
     by = c("groupId")) %>%
-    rowwise() %>%
-    mutate(!!rlang::sym(quant_var) := rnorm(1, mean = !!rlang::sym(quant_var)+1, sd = imputation_sd))
+    dplyr::rowwise() %>%
+    dplyr::mutate(!!rlang::sym(quant_var) := stats::rnorm(1, mean = !!rlang::sym(quant_var)+1, sd = imputation_sd))
   
   ## merge measured peaks with imputed peaks
   completed_peaks <- dplyr::bind_rows(
@@ -125,6 +146,15 @@ impute_missing_peaks <- function(mzroll_list,
 #'@return triple omic data with imputed missing peaks
 #'
 #' @examples
+#' library(dplyr)
+#' 
+#' lod_values <- nplug_mzroll_augmented[["measurements"]] %>%
+#' dplyr::select(groupId, log2_abundance) %>%
+#' dplyr::distinct(groupId, .keep_all = TRUE)
+#' 
+#' mzroll_list <- nplug_mzroll_augmented
+#' mzroll_list$measurements <- mzroll_list$measurements %>% filter(groupId != 2 & sampleId != 1)
+#' 
 #' mzroll_list_imputed <- fill_in_missing_peaks(mzroll_list, lod_values, "log2_abundance", 0.15)
 #' mzroll_list_imputed <- fill_in_missing_peaks(mzroll_list, 12, "log2_abundance")
 #'
@@ -134,8 +164,7 @@ fill_in_missing_peaks <- function(mzroll_list,
                                   quant_var = "log2_abundance",
                                   imputation_sd = 0.15) {
   if (is.data.frame(fill_values)) {
-    stopifnot(colnames(fill_values) %in% c("groupId", rlang::sym(quant_var)))
-    output <- impute_missing_peaks(mzroll_list, lod_values, quant_var, imputation_sd)
+    output <- impute_missing_peaks(mzroll_list, fill_values, quant_var, imputation_sd)
   } else if (is.numeric(fill_values)) {
     output <- floor_peaks(mzroll_list, fill_values, quant_var)
   } else {
diff --git a/man/fill_in_missing_peaks.Rd b/man/fill_in_missing_peaks.Rd
index dee2c94..8c1e1ce 100644
--- a/man/fill_in_missing_peaks.Rd
+++ b/man/fill_in_missing_peaks.Rd
@@ -28,6 +28,15 @@ If \code{fill_values} is a data frame, this function calls \code{impute_missing_
 If it is a numeric vector, 'this function calls \code{floor_peaks()}. Other types are not currently supported
 }
 \examples{
+library(dplyr)
+
+lod_values <- nplug_mzroll_augmented[["measurements"]] \%>\%
+dplyr::select(groupId, log2_abundance) \%>\%
+dplyr::distinct(groupId, .keep_all = TRUE)
+
+mzroll_list <- nplug_mzroll_augmented
+mzroll_list$measurements <- mzroll_list$measurements \%>\% filter(groupId != 2 & sampleId != 1)
+
 mzroll_list_imputed <- fill_in_missing_peaks(mzroll_list, lod_values, "log2_abundance", 0.15)
 mzroll_list_imputed <- fill_in_missing_peaks(mzroll_list, 12, "log2_abundance")
 
diff --git a/man/impute_missing_peaks.Rd b/man/impute_missing_peaks.Rd
index c3e9c3a..a735c6b 100644
--- a/man/impute_missing_peaks.Rd
+++ b/man/impute_missing_peaks.Rd
@@ -27,6 +27,15 @@ triple omic data with imputed missing peaks
 Impute missing peaks with provided feature-level imputation values
 }
 \examples{
+library(dplyr)
+
+lod_values <- nplug_mzroll_augmented[["measurements"]] \%>\%
+dplyr::select(groupId, log2_abundance) \%>\%
+dplyr::distinct(groupId, .keep_all = TRUE)
+
+mzroll_list <- nplug_mzroll_augmented
+mzroll_list$measurements <- mzroll_list$measurements \%>\% filter(groupId != 2 & sampleId != 1)
+
 mzroll_list_imputed <- impute_missing_peaks(mzroll_list, lod_values, "log2_abundance", 0.15)
 
 }
diff --git a/man/plot_volcano.Rd b/man/plot_volcano.Rd
index 862854b..df6fba4 100644
--- a/man/plot_volcano.Rd
+++ b/man/plot_volcano.Rd
@@ -28,13 +28,15 @@ a grob
 Volcano plot
 }
 \examples{
+library(dplyr)
 regression_significance <- diffex_mzroll(
-  nplug_mzroll_normalized,
-  "normalized_log2_abundance",
-  "limitation + limitation:DR + 0",
-  FDR_cutoff = 0.01,
-  feature_labels = c("UDP", "Lactate", "Serine")
-)
+nplug_mzroll_normalized,
+"normalized_log2_abundance",
+"limitation + limitation:DR + 0") \%>\%
+dplyr::left_join(
+nplug_mzroll_normalized$features \%>\% select(groupId, compoundName),
+by = "groupId")
+
+plot_volcano(regression_significance, 10, 0.1, c("Ribose-P", "acetyl-CoA", "ATP"))
 
-plot_volcano(regression_significance, 10, 0.1)
 }
diff --git a/vignettes/NPLUG.Rmd b/vignettes/NPLUG.Rmd
index 1dc7a87..566f800 100644
--- a/vignettes/NPLUG.Rmd
+++ b/vignettes/NPLUG.Rmd
@@ -287,7 +287,7 @@ Two visualizations that are particularly useful for EDA are heatmaps (I'm sure y
 
 ```{r static_eda, fig.height = 8, fig.width = 8}
 samples_with_pcs <- final_processed_data %>%
-  romic::add_pca_loadings(value_var = "normalized_log2_abundance", npcs = 5)
+  romic::add_pcs(value_var = "normalized_log2_abundance", npcs = 5)
 
 romic::plot_bivariate(
   samples_with_pcs$samples,

From 41ffe7736cca9b169058ca89cda2977d1817ff1d Mon Sep 17 00:00:00 2001
From: delfarahalireza <delfarah@calicolabs.com>
Date: Tue, 21 Nov 2023 15:45:42 -0800
Subject: [PATCH 7/7] fixing validation check in impute_missing_peaks()

---
 R/mutate_mzroll_list.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/R/mutate_mzroll_list.R b/R/mutate_mzroll_list.R
index fe71e5a..a066244 100644
--- a/R/mutate_mzroll_list.R
+++ b/R/mutate_mzroll_list.R
@@ -93,7 +93,7 @@ impute_missing_peaks <- function(mzroll_list,
   
   features <- mzroll_list$features %>% dplyr::select(groupId)
   
-  if(!all(features$groupId %in% lod_values$groupId) | nrow(features) != nrow(lod_values)) {
+  if(!all(features$groupId %in% lod_values$groupId)) {
     stop("groupId values of lod_value table and feature table of triple omic data must match")
   }