Mol-subtyping using pathology_free_text_diagnosis reported subtype (M…

…eningioma) update (#1047)

* add path_free_text subtypes to mol subtype integrate

* rerun bindrows

* rerun harm_dx meningioma

* rerun to add tumor_descriptor

* Update analyses/molecular-subtyping-pathology/pathology_free_text-subtyping-meningioma.Rmd

Co-authored-by: Jo Lynne Rokita <jharenza@gmail.com>

* adding order to output

* Update run-subtyping-aggregation.sh

* Update run-subtyping-aggregation.sh

* Update pathology_free_text-subtyping-meningioma.Rmd

Co-authored-by: Jo Lynne Rokita <jharenza@gmail.com>

analyses/molecular-subtyping-integrate/01-integrate-subtyping.Rmd

@@ -49,11 +49,11 @@ base_histology <- read_tsv(file.path(data_dir,"pbta-histologies-base.tsv"),
 
 ### Read molecular-subtyping-pathology results
 
-Reading molecular_subtype, integrated_diagnosis, short_histology, broad_histology and Notes from `compiled_molecular_subtypes_with_clinical_pathology_feedback.tsv` 
+Reading molecular_subtype, integrated_diagnosis, short_histology, broad_histology and Notes from `compiled_molecular_subtypes_with_clinical_pathology_feedback_and_report_info.tsv` 
 
 ```{r}
 
-compiled_subtyping<-read_tsv(file.path("..", "molecular-subtyping-pathology", "results", "compiled_molecular_subtypes_with_clinical_pathology_feedback.tsv"))
+compiled_subtyping<-read_tsv(file.path("..", "molecular-subtyping-pathology", "results", "compiled_molecular_subtypes_with_clinical_pathology_feedback_and_report_info.tsv"))
 
 ```
 
@@ -208,5 +208,6 @@ histology %>%
          -broad_histology.subtyped,
          - short_histology.base,
          -short_histology.subtyped) %>%
+  arrange(Kids_First_Biospecimen_ID) %>%
   write_tsv("results/pbta-histologies.tsv")
 ```

analyses/molecular-subtyping-pathology/clinical-subtyping-meningioma.Rmd → analyses/molecular-subtyping-pathology/pathology_free_text-subtyping-meningioma.Rmd

@@ -20,13 +20,17 @@ We will use this notebook to do so; see [#993](https://github.com/AlexsLemonade/
 ```{r}
 # Pipes
 library(magrittr)
+library(tidyverse)
 ```
 
 ### Input
 
 ```{r}
 data_dir <- file.path("..", "..", "data")
-histologies_file <- file.path(data_dir, "pbta-histologies.tsv")
+results_dir <- file.path("results")
+histologies_file <- file.path(data_dir, "pbta-histologies-base.tsv")
+compiled_mol_subtypes_pathology_clinical_file <- file.path(results_dir,
+                         "compiled_molecular_subtypes_with_clinical_pathology_feedback.tsv")
 ```
 
 ### Output
@@ -37,13 +41,14 @@ if (!dir.exists(results_dir)) {
   dir.create(results_dir)
 }
 
-output_file <- file.path(results_dir, "meningioma_subtypes.tsv")
+output_file <- file.path(results_dir, "compiled_molecular_subtypes_with_clinical_pathology_feedback_and_report_info.tsv")
 ```
 
 ## Read in data
 
 ```{r}
 histologies_df <- readr::read_tsv(histologies_file, guess_max = 10000)
+compiled_mol_subtypes_pathology_clinical_df <- read_tsv(compiled_mol_subtypes_pathology_clinical_file, guess = 10000)
 ```
 
 ### Display `pathology_free_text_diagnosis` values
@@ -77,11 +82,10 @@ meningioma_df <- histologies_df %>%
   dplyr::select(Kids_First_Biospecimen_ID,
                 Kids_First_Participant_ID,
                 sample_id,
-                pathology_diagnosis,
+                tumor_descriptor,
                 pathology_free_text_diagnosis,
                 broad_histology,
-                short_histology,
-                harmonized_diagnosis) %>%
+                short_histology) %>%
   # To smooth the way for string detection for the pathology free text, we
   # add a column where all of the text is converted to lowercase
   dplyr::mutate(pathology_free_text_dx_lower = 
@@ -100,21 +104,29 @@ meningioma_df <- histologies_df %>%
       # This will be true when none of the conditions (e.g., strings are 
       # detected) above
       TRUE ~ "Meningioma"
-    )
+    ),
+    Notes = "Updated via OpenPBTA subtyping from pathology_free_text_diagnosis"
   ) %>%
   # Drop the column we added for convenience of string detection
-  dplyr::select(-pathology_free_text_dx_lower)
+  # and to format to match compiled_mol_subtypes_pathology_clinical_df
+  dplyr::select(-pathology_free_text_dx_lower,
+                -pathology_free_text_diagnosis)
 ```
 
 Write to file!
+Since Meningioma was not subtyped as part of OpenPBTA molecular subtying we can append the meningioma results to `compiled_mol_subtypes_pathology_clinical_df`.
 
 ```{r}
-readr::write_tsv(meningioma_df, output_file)
+
+meningioma_df %>%
+  bind_rows(compiled_mol_subtypes_pathology_clinical_df) %>%
+  arrange(Kids_First_Biospecimen_ID) %>%
+  readr::write_tsv(output_file)
+
 ```
 
 ## Session Info
 
 ```{r}
 sessionInfo()
 ```
-