Datasets Merged, Build works, Validate works

Author: jjacobson95

build/mpnst/00_sample_gen.R

@@ -1,5 +1,4 @@
-# This script generate a new sample table based on pervious beatAML improved sample ID
-# It will take the maximum value of beatAML improved sample ID and continue from ID count from there
+# This script generate a new sample table based on previous dataset's sample file (taking the max improve_sample_id)
 # Load required libraries
 library(data.table)
 library(synapser)

build/mpnst/01_combined_omics.R

@@ -0,0 +1,250 @@
+#!/usr/bin/env Rscript
+
+# Combined MPNST & MPNST-PDX Data Extraction Script
+# This script unifies data extraction for PDX, Tumor, and Xenograft-Derived Organoid samples.
+
+# Load required libraries
+library(data.table)
+library(synapser)
+library(dplyr)
+library(tidyr)
+
+# Retrieve command line arguments
+args <- commandArgs(trailingOnly = TRUE)
+if (length(args) < 3) {
+  stop("Usage: Rscript 01_combined_omics.R <PAT> <samples.csv> <genes.csv>", call. = FALSE)
+}
+PAT      <- args[1]
+samples  <- args[2]
+genes    <- args[3]
+
+# Log in to Synapse
+token <- PAT
+synLogin(authToken = token)
+
+# Read sample mapping and gene mapping
+samples_df <- fread(samples) %>%
+  select(improve_sample_id, common_name, model_type) %>%
+  distinct()
+genes_df <- fread(genes)
+
+# Subset by model type
+pdx_samps   <- filter(samples_df, model_type == "patient derived xenograft")
+tumor_samps<- filter(samples_df, model_type == "tumor")
+mt_samps    <- filter(samples_df, model_type == "xenograft derived organoid")  # These end up being the same as pdx_samps in the manifest.
+
+# Retrieve manifest table from Synapse
+manifest <- synTableQuery("select * from syn53503360")$asDataFrame() %>%
+  rename(common_name = Sample)
+
+print("manifest")
+print(manifest)
+
+# Build sample tables
+pdx_data <- manifest %>%
+  select(common_name, starts_with("PDX")) %>%
+  left_join(pdx_samps, by = "common_name") %>%
+  select(improve_sample_id, common_name, model_type,
+         RNASeq = PDX_RNASeq,
+         Mutations = PDX_Somatic_Mutations,
+         CopyNumber = PDX_CNV,
+         Proteomics = PDX_Proteomics) %>%
+  filter(!is.na(improve_sample_id))
+
+
+tumor_data <- manifest %>%
+  select(common_name, starts_with("Tumor")) %>%
+  left_join(tumor_samps, by = "common_name") %>%
+  select(improve_sample_id, common_name, model_type,
+         RNASeq = Tumor_RNASeq,
+         Mutations = Tumor_Somatic_Mutations,
+         CopyNumber = Tumor_CNV) %>%
+  mutate(Proteomics = "") %>%
+  filter(!is.na(improve_sample_id))
+
+mt_data <- manifest %>%                     #Note, this is the same as pdx_data but I think we default to "xenograft derived organoid" if present.
+  select(common_name, starts_with("PDX")) %>%
+  left_join(mt_samps, by = "common_name") %>%
+  select(improve_sample_id, common_name, model_type,
+         RNASeq = PDX_RNASeq,
+         Mutations = PDX_Somatic_Mutations,
+         CopyNumber = PDX_CNV,
+         Proteomics = PDX_Proteomics) %>%
+  filter(!is.na(improve_sample_id))
+
+# Combine all sample tables
+dcombined <- bind_rows(pdx_data, tumor_data, mt_data) %>% distinct()
+print("dcombined:")
+print(dcombined)
+
+# Helper to assign study label based on model_type
+study_label <- function(type) {
+  case_when(
+    type == "patient derived xenograft"     ~ "MPNST PDX",
+    type == "tumor"                          ~ "MPNST Tumor",
+    type == "xenograft derived organoid"     ~ "MPNST PDX MT",
+    TRUE                                       ~ "MPNST"
+  )
+}
+
+# Helper to pick metadata based on sample ID and column
+pick_meta <- function(id, column) {
+  # column  {"Proteomics","RNASeq","Mutations","CopyNumber"}
+  if (any(tumor_data[[column]] == id, na.rm = TRUE)) {
+    sdf <- tumor_data %>% filter(.data[[column]] == id) %>% slice(1)
+  } else if (any(mt_data[[column]] == id, na.rm = TRUE)) {
+    sdf <- mt_data    %>% filter(.data[[column]] == id) %>% slice(1)
+  } else if (any(pdx_data[[column]] == id, na.rm = TRUE)) {
+    sdf <- pdx_data   %>% filter(.data[[column]] == id) %>% slice(1)
+  } else {
+    return(NULL)
+  }
+  list(
+    sample_id  = sdf$improve_sample_id,
+    model_type = sdf$model_type
+  )
+}
+
+# Safe extraction: only return non-empty data frames
+i_safe_extract <- function(df, sample_id, source_val, study_val) {
+  if (is.null(df) || nrow(df) == 0) return(NULL)
+  df$improve_sample_id <- sample_id
+  df$source            <- source_val
+  df$study             <- study_val
+  df
+}
+
+# 1) Proteomics
+proteomics_list <- lapply(
+  setdiff(dcombined$Proteomics, c("", NA, "NA")),
+  function(id) {
+    meta <- pick_meta(id, "Proteomics")
+    if (is.null(meta)) return(NULL)
+
+    df <- tryCatch(
+      fread(synGet(id)$path) %>%
+        rename(gene_symbol = Gene) %>%
+        left_join(genes_df, by = "gene_symbol") %>%
+        select(entrez_id, proteomics = logRatio) %>%
+        filter(!is.na(entrez_id), proteomics != 0) %>%
+        distinct(),
+      error = function(e) NULL
+    )
+    i_safe_extract(
+      df,
+      meta$sample_id,
+      "NF Data Portal",
+      study_label(meta$model_type)
+    )
+  }
+)
+proteomics <- bind_rows(proteomics_list)
+fwrite(proteomics, file.path("/tmp", "mpnst_proteomics.csv"))
+message("Wrote combined proteomics")
+
+
+# 2) Transcriptomics (PDX, Tumor, and Organoid / MT which comes from PDX..)
+transcriptomics_list <- lapply(
+  setdiff(dcombined$RNASeq, c("", NA, "NA")),
+  function(id) {
+    meta <- pick_meta(id, "RNASeq")
+    if (is.null(meta)) return(NULL)
+
+    df <- tryCatch({
+      fread(synGet(id)$path) %>%
+        separate(Name, into = c("other_id","vers"), sep = "\\.") %>%
+        select(-vers) %>%
+        left_join(genes_df) %>%
+        select(entrez_id, transcriptomics = TPM) %>%
+        filter(!is.na(entrez_id), transcriptomics != 0) %>%
+        distinct()
+    }, error = function(e) NULL)
+
+    i_safe_extract(
+      df,
+      meta$sample_id,
+      "NF Data Portal",
+      study_label(meta$model_type)
+    )
+  }
+)
+transcriptomics <- bind_rows(transcriptomics_list)
+fwrite(transcriptomics, file.path("/tmp", "mpnst_transcriptomics.csv"))
+message("Wrote combined transcriptomics")
+
+
+# 3) Mutations (WES)
+wes_list <- lapply(
+  setdiff(dcombined$Mutations, c("", NA, "NA")),
+  function(id) {
+    meta <- pick_meta(id, "Mutations")
+    if (is.null(meta)) return(NULL)
+
+    clean_id <- gsub('[\"\\[\\]]', '', id)
+    df <- tryCatch(
+      fread(synGet(clean_id)$path) %>%
+        select(entrez_id = Entrez_Gene_Id,
+               mutation               = HGVSc,
+               variant_classification = Variant_Classification) %>%
+        filter(entrez_id %in% genes_df$entrez_id) %>%
+        distinct(),
+      error = function(e) NULL
+    )
+
+    i_safe_extract(
+      df,
+      meta$sample_id,
+      "NF Data Portal",
+      study_label(meta$model_type)
+    )
+  }
+)
+wes <- bind_rows(wes_list)
+fwrite(wes, file.path("/tmp", "mpnst_mutations.csv"))
+message("Wrote combined mutations")
+
+
+# 4) Copy Number Variation (CNV)
+cnv_list <- lapply(
+  setdiff(dcombined$CopyNumber, c("", NA, "NA")),
+  function(id) {
+    meta <- pick_meta(id, "CopyNumber")
+    if (is.null(meta)) return(NULL)
+
+    clean_id <- gsub('[\"\\[\\]]', '', id)
+    raw <- tryCatch(fread(synGet(clean_id)$path), error = function(e) NULL)
+    if (is.null(raw)) return(NULL)
+
+    df_long <- raw %>%
+      separate_rows(gene, sep = ",") %>%
+      rename(gene_symbol = gene) %>%
+      left_join(genes_df, by = "gene_symbol") %>%
+      filter(!is.na(entrez_id)) %>%
+      select(entrez_id, log2) %>%
+      distinct() %>%
+      mutate(copy_number = 2^log2) %>%
+      select(-log2)
+
+    df <- df_long %>%
+      mutate(copy_call = case_when(
+        copy_number < 0.5210507 ~ "deep del",
+        copy_number < 0.7311832 ~ "het loss",
+        copy_number < 1.214125  ~ "diploid",
+        copy_number < 1.422233  ~ "gain",
+        TRUE                    ~ "amp"
+      ))
+
+    i_safe_extract(
+      df,
+      meta$sample_id,
+      "NF Data Portal",
+      study_label(meta$model_type)
+    )
+  }
+)
+cnv <- bind_rows(cnv_list)
+fwrite(cnv, file.path("/tmp", "mpnst_copy_number.csv"))
+message("Wrote combined copy number")
+
+
+message("All combined data files created.")