From e9c4f2343765d61684cee33a525933708f573841 Mon Sep 17 00:00:00 2001
From: Anh Nguyet Vu <anngvu@gmail.com>
Date: Tue, 19 Mar 2024 17:18:53 -0600
Subject: [PATCH] Add prelim script

---
 manage-sample-ids/README.md |  4 ++++
 manage-sample-ids/extract.R | 27 +++++++++++++++++++++++++++
 2 files changed, 31 insertions(+)
 create mode 100644 manage-sample-ids/README.md
 create mode 100644 manage-sample-ids/extract.R

diff --git a/manage-sample-ids/README.md b/manage-sample-ids/README.md
new file mode 100644
index 0000000..f6e6f3a
--- /dev/null
+++ b/manage-sample-ids/README.md
@@ -0,0 +1,4 @@
+## Sample ID Management
+
+Collection of utility scripts that help with managing sample entities for the BEN Collaborative Project / Breast Atlas in Synapse. 
+There is potential functionality for extracting ids, inferring linkage of existing sample entities with data files, etc., that may be more automated eventually.
diff --git a/manage-sample-ids/extract.R b/manage-sample-ids/extract.R
new file mode 100644
index 0000000..1eed5fb
--- /dev/null
+++ b/manage-sample-ids/extract.R
@@ -0,0 +1,27 @@
+library(synapser)
+
+synapser::synLogin(authToken = Sys.getenv("SYNAPSE_AUTH_TOKEN"))
+
+# It is easiest to use a fileview to get all nested children in "Samples"
+SampleTracking <- "syn52225331"
+
+tb <- synapser::synTableQuery(glue::glue("select id from {SampleTracking}"))
+ids <- as.data.frame(tb)$id
+
+asPatientID <- function(s) regmatches(s, regexpr("[A-Z]+[0-9]?-?[0-9]+", s))
+
+errors <- c()
+
+for(i in ids) {
+  
+  entity <- synapser::synGet(i, downloadFile = FALSE)
+  pID <- asPatientID(entity$properties$name)
+  entity$annotations$PatientID <- pID
+  
+  # For now, let's just use the entity name as sample name
+  entity$annotations$SampleID <- entity$properties$name
+  tryCatch(synapser::synStore(entity), error = function(e) { cat("Skipped ", i, "\n"); errors <- append(errors, i) })
+}
+
+
+