docs(vignette): Integrate MSstats workflow into vignette and test data

Vitek-Lab · Aug 27, 2024 · a09b9bd · a09b9bd
1 parent 8c5e873
commit a09b9bd
Show file tree

Hide file tree

Showing 13 changed files with 61 additions and 23 deletions.
diff --git a/R/getSubnetworkFromIndra.R b/R/getSubnetworkFromIndra.R
@@ -11,7 +11,7 @@
 #'
 #' @examples
 #' input <- data.table::fread(system.file(
-#'     "processed_data/groupComparisonModel.csv",
+#'     "extdata/groupComparisonModel.csv",
 #'     package = "MSstatsBioNet"
 #' ))
 #' # subnetwork = getSubnetworkFromIndra(input, pvalue_cutoff = 0.05)

diff --git a/R/visualizeNetworks.R b/R/visualizeNetworks.R
@@ -13,7 +13,7 @@
 #'
 #' @examples
 #' input <- data.table::fread(system.file(
-#'     "processed_data/groupComparisonModel.csv",
+#'     "extdata/groupComparisonModel.csv",
 #'     package = "MSstatsBioNet"
 #' ))
 #' # subnetwork = getSubnetworkFromIndra(input)

diff --git a/inst/extdata/groupComparisonModel.csv b/inst/extdata/groupComparisonModel.csv
@@ -0,0 +1,8 @@
+"Protein","Label","log2FC","SE","Tvalue","DF","pvalue","adj.pvalue","issue","MissingPercentage","ImputationPercentage","HgncId"
+"B9A064","neg vs pos",0.289168813227902,0.0290298430266267,9.96108773177556,2,0.00992843836713075,0.0694990685699153,NA,0,0,"38476"
+"O00391","neg vs pos",0.61514367352039,0.607915303579319,1.01189042272585,1,0.496237573382871,0.496237573382871,NA,0.25,0,"9756"
+"O14818","neg vs pos",-0.351724166825181,0.180293183487185,-1.95084561724532,2,0.190360070677825,0.333130123686193,NA,0.208333333333333,0.208333333333333,"9536"
+"O43598","neg vs pos",0.444322024324174,0.222001037180599,2.00144120931613,2,0.18330743657428,0.333130123686193,NA,0.1875,0.1875,"21218"
+"O43707","neg vs pos",-0.678628567390453,0.196242424527707,-3.45811344832135,2,0.0744090385980984,0.260431635093345,NA,0.140625,0.140625,"166"
+"P16050","neg vs pos",0.239036560821488,0.213802694760769,1.11802407864388,2,0.379829709925256,0.443134661579466,NA,0.134615384615385,0.134615384615385,"11390"
+"P84243","neg vs pos",-1.03311234546861,0.833093604461367,-1.24009155746258,2,0.3406963949584,0.443134661579466,NA,0.125,0.125,"4765"
diff --git a/inst/extdata/indraResponse.rds b/inst/extdata/indraResponse.rds
diff --git a/inst/processed_data/subnetwork.rds → inst/extdata/subnetwork.rds b/inst/processed_data/subnetwork.rds → inst/extdata/subnetwork.rds
diff --git a/inst/processed_data/groupComparisonModel.csv b/inst/processed_data/groupComparisonModel.csv
diff --git a/inst/processed_data/indraResponse.rds b/inst/processed_data/indraResponse.rds
diff --git a/inst/script/groupComparisonModel.R b/inst/script/groupComparisonModel.R
@@ -0,0 +1,31 @@
+# Script to create groupComparisonModel.csv in extdata folder
+library(MSstatsConvert)
+input = system.file("tinytest/raw_data/Metamorpheus/AllQuantifiedPeaks.tsv", 
+                    package = "MSstatsConvert")
+input = data.table::fread(input)
+annot = system.file("tinytest/raw_data/Metamorpheus/Annotation.tsv", 
+                    package = "MSstatsConvert")
+annot = data.table::fread(annot)
+msstats_imported = MetamorpheusToMSstatsFormat(input, annotation = annot)
+head(msstats_imported)
+
+library(MSstats)
+QuantData = dataProcess(msstats_imported, use_log_file = FALSE)
+groupComparisonResult = groupComparison(contrast.matrix="pairwise", 
+                                        data=QuantData, 
+                                        use_log_file = FALSE)
+
+uniprot_to_hgnc_mapping = c(
+    "B9A064" = "38476",
+    "O00391" = "9756",
+    "O14818" = "9536",
+    "O43598" = "21218",
+    "O43707" = "166",
+    "P16050" = "11390",
+    "P84243" = "4765"
+)
+groupComparisonResult$ComparisonResult$HgncId = uniprot_to_hgnc_mapping[
+    groupComparisonResult$ComparisonResult$Protein
+]
+
+write.csv(groupComparisonResult$ComparisonResult, "groupComparison.csv", row.names = FALSE)
diff --git a/man/getSubnetworkFromIndra.Rd b/man/getSubnetworkFromIndra.Rd
diff --git a/man/visualizeNetworks.Rd b/man/visualizeNetworks.Rd
diff --git a/tests/testthat/test-getSubnetworkFromIndra.R b/tests/testthat/test-getSubnetworkFromIndra.R
@@ -1,23 +1,23 @@
 test_that("getSubnetworkFromIndra works correctly", {
     input <- data.table::fread(
-        system.file("processed_data/groupComparisonModel.csv", package = "MSstatsBioNet")
+        system.file("extdata/groupComparisonModel.csv", package = "MSstatsBioNet")
     )
     local_mocked_bindings(.callIndraCogexApi = function(x) {
-        return(readRDS(system.file("processed_data/indraResponse.rds", package = "MSstatsBioNet")))
+        return(readRDS(system.file("extdata/indraResponse.rds", package = "MSstatsBioNet")))
     })
     subnetwork <- getSubnetworkFromIndra(input)
-    expect_equal(nrow(subnetwork$nodes), 4)
-    expect_equal(nrow(subnetwork$edges), 16)
+    expect_equal(nrow(subnetwork$nodes), 7)
+    expect_equal(nrow(subnetwork$edges), 2)
 })
 
 test_that("getSubnetworkFromIndra with pvalue filter works correctly", {
     input <- data.table::fread(
-        system.file("processed_data/groupComparisonModel.csv", package = "MSstatsBioNet")
+        system.file("extdata/groupComparisonModel.csv", package = "MSstatsBioNet")
     )
     local_mocked_bindings(.callIndraCogexApi = function(x) {
-        return(readRDS(system.file("processed_data/indraResponse.rds", package = "MSstatsBioNet")))
+        return(readRDS(system.file("extdata/indraResponse.rds", package = "MSstatsBioNet")))
     })
-    subnetwork <- getSubnetworkFromIndra(input, pvalue_cutoff = 0.05)
-    expect_equal(nrow(subnetwork$nodes), 3)
-    expect_equal(nrow(subnetwork$edges), 16)
+    subnetwork <- getSubnetworkFromIndra(input, pvalue_cutoff = 0.45)
+    expect_equal(nrow(subnetwork$nodes), 6)
+    expect_equal(nrow(subnetwork$edges), 2)
 })
diff --git a/tests/testthat/test-visualizeNetworks.R b/tests/testthat/test-visualizeNetworks.R
@@ -1,5 +1,5 @@
 test_that("visualizeNetworks works correctly", {
-    input <- readRDS(system.file("processed_data/subnetwork.rds",
+    input <- readRDS(system.file("extdata/subnetwork.rds",
         package = "MSstatsBioNet"
     ))
 
@@ -33,7 +33,7 @@ test_that("visualizeNetworks works correctly", {
 
 
 test_that("visualizeNetworks with p-value and logFC constraints works", {
-    input <- readRDS(system.file("processed_data/subnetwork.rds",
+    input <- readRDS(system.file("extdata/subnetwork.rds",
         package = "MSstatsBioNet"
     ))
 

diff --git a/vignettes/MSstatsBioNet.Rmd b/vignettes/MSstatsBioNet.Rmd
@@ -36,14 +36,15 @@ input = data.table::fread(input)
 annot = system.file("tinytest/raw_data/Metamorpheus/Annotation.tsv", 
                                 package = "MSstatsConvert")
 annot = data.table::fread(annot)
-msstats_imported = MetamorpheusToMSstatsFormat(input, annotation = annot)
+msstats_imported = MetamorpheusToMSstatsFormat(input, annotation = annot, 
+                                               use_log_file = FALSE)
 head(msstats_imported)
 ```
 
 # MSstats Process and GroupComparison
 
 ```{r}
-library(MSstatsConvert)
+library(MSstats)
 QuantData = dataProcess(msstats_imported, use_log_file = FALSE)
 groupComparisonResult = groupComparison(contrast.matrix="pairwise", 
                                         data=QuantData, 
@@ -60,6 +61,9 @@ containing HGNC IDs.  We can use the following mapping to convert from Uniprot
 to HGNC ID.  This mapping was obtained from this 
 [Python uniprot client](https://indra.readthedocs.io/en/docstrings/modules/databases/index.html#indra.databases.hgnc_client.get_hgnc_id)
 
+In the future, this uniprot client will be added as a separate function to allow
+users to perform this conversion in R in a streamlined way.
+
 ```{r}
 uniprot_to_hgnc_mapping = c(
     "B9A064" = "38476",
@@ -91,7 +95,7 @@ This package is distributed under the [Artistic-2.0](https://opensource.org/lice
 
 ## Visualize Networks
 
-```{r proteinNetworkDiscovery, eval=FALSE}
+```{r visualizeNetwork, eval=FALSE}
 visualizeNetworks(subnetwork$nodes, subnetwork$edges)
 ```