From 86e04962ebb67dd49c61f262758b948e4a2de3c1 Mon Sep 17 00:00:00 2001
From: Tim O'Donnell <timodonnell@gmail.com>
Date: Thu, 12 Dec 2019 13:48:43 -0500
Subject: [PATCH] update

---
 .../models_class1_pan_refined/GENERATE.sh         | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/downloads-generation/models_class1_pan_refined/GENERATE.sh b/downloads-generation/models_class1_pan_refined/GENERATE.sh
index 3cae8f14..4674cdc2 100755
--- a/downloads-generation/models_class1_pan_refined/GENERATE.sh
+++ b/downloads-generation/models_class1_pan_refined/GENERATE.sh
@@ -89,11 +89,16 @@ fi
 
 # ********************************************************
 echo "Beginning production run"
-time python make_multiallelic_training_data.py \
-    --hits "$(mhcflurry-downloads path data_mass_spec_annotated)/annotated_ms.csv.bz2" \
-    --expression "$(mhcflurry-downloads path data_curated)/rna_expression.csv.bz2" \
-    --decoys-per-hit 1 \
-    --out train.multiallelic.csv
+if [ -f "$SCRIPT_DIR/train.multiallelic.csv" ]; then
+    echo "Using existing multiallelic train data."
+    cp "$SCRIPT_DIR/train.multiallelic.csv" .
+else
+    time python make_multiallelic_training_data.py \
+        --hits "$(mhcflurry-downloads path data_mass_spec_annotated)/annotated_ms.csv.bz2" \
+        --expression "$(mhcflurry-downloads path data_curated)/rna_expression.csv.bz2" \
+        --decoys-per-hit 1 \
+        --out train.multiallelic.csv
+fi
 
 ALLELE_LIST=$(bzcat "$MONOALLELIC_TRAIN" | cut -f 1 -d , | grep -v allele | uniq | sort | uniq)
 ALLELE_LIST+=$(cat train.multiallelic.csv | cut -f 7 -d , | gerp -v hla | uniq | tr ' ' '\n' | sort | uniq)
-- 
GitLab