From 3fd911a277d5e78c84c9f3f713a19de928bc9ba2 Mon Sep 17 00:00:00 2001
From: Marcel Levstek <62072754+marcellevstek@users.noreply.github.com>
Date: Fri, 20 Dec 2024 11:55:48 +0100
Subject: [PATCH] fixup! Make changes to ``gtf-to-bed`` process

---
 .../support_processors/gtf_to_bed.py          | 23 +++++++++----------
 .../processes/test_support_processors.py      |  2 ++
 2 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/resolwe_bio/processes/support_processors/gtf_to_bed.py b/resolwe_bio/processes/support_processors/gtf_to_bed.py
index 1b57081e1..b6d084173 100644
--- a/resolwe_bio/processes/support_processors/gtf_to_bed.py
+++ b/resolwe_bio/processes/support_processors/gtf_to_bed.py
@@ -15,7 +15,7 @@
 
 
 class GTFtoBED(Process):
-    """GTF to BED conversion.
+    """GTF to BED conversion for predefined genes and feature types.
 
     Note that this process only works with ENSEMBL annotations.
     """
@@ -100,7 +100,7 @@ class Input:
             "geneset",
             label="Gene set",
             description="Gene set to use for filtering.",
-            required=False,
+            required=True,
         )
 
         canonical_transcripts = DataField(
@@ -181,17 +181,16 @@ def run(self, inputs, outputs):
         gtf = gtf[gtf["source"].isin(inputs.annotation_source)]
         gtf = gtf[gtf["feature_type"] == feature_type]
 
-        if inputs.geneset:
-            if inputs.annotation.output.species != inputs.geneset.output.species:
-                self.error(
-                    "Gene set data object species does not match the annotation species."
-                )
-            geneset = pd.read_csv(
-                inputs.geneset.output.geneset.path,
-                delimiter="\t",
-                names=["ID"],
+        if inputs.annotation.output.species != inputs.geneset.output.species:
+            self.error(
+                "Species of the gene set data object does not match the species of the annotation data object."
             )
-            gtf = gtf[gtf["gene_id"].isin(geneset["ID"])]
+        geneset = pd.read_csv(
+            inputs.geneset.output.geneset.path,
+            delimiter="\t",
+            names=["ID"],
+        )
+        gtf = gtf[gtf["gene_id"].isin(geneset["ID"])]
 
         if inputs.canonical_transcripts and not feature_type == "gene":
             if (
diff --git a/resolwe_bio/tests/processes/test_support_processors.py b/resolwe_bio/tests/processes/test_support_processors.py
index 09783c676..2e7a64d30 100644
--- a/resolwe_bio/tests/processes/test_support_processors.py
+++ b/resolwe_bio/tests/processes/test_support_processors.py
@@ -1789,6 +1789,7 @@ def test_gtf_to_bed(self):
             "gtf-to-bed",
             {
                 "annotation": gtf.id,
+                "geneset": geneset.id,
                 "annotation_field": "gene_name",
             },
         )
@@ -1798,6 +1799,7 @@ def test_gtf_to_bed(self):
             "gtf-to-bed",
             {
                 "annotation": gtf.id,
+                "geneset": geneset.id,
                 "annotation_field": "gene_id_feature_id",
                 "feature_type": "exon",
             },