Merge pull request #1 from mskcc/feature/metafusion_parser

Add CFF parser with tests

README.md

@@ -156,6 +156,7 @@ You can provide as input output files from fusion-finding algorithms. Currently
 * MapSplice (only if --gene-gtf specified)
 * [STAR-Fusion](https://github.com/STAR-Fusion/STAR-Fusion)
 * TopHat-Fusion
+* [Common Fusion Format](https://github.com/ccmbioinfo/MetaFusion/wiki/metafusion-file-formats#cff-fields)
 
 
 

agfusion/model.py

@@ -691,6 +691,8 @@ def save_tables(self, out_dir="."):
                 "3'_gene",
                 "5'_transcript",
                 "3'_transcript",
+                "5'_breakpoint",
+                "3'_breakpoint",
                 "5'_strand",
                 "3'_strand",
                 "5'_transcript_biotype",
@@ -720,6 +722,8 @@ def save_tables(self, out_dir="."):
                     transcript.gene3prime.gene.gene_name,
                     transcript.transcript1.id,
                     transcript.transcript2.id,
+                    transcript.gene5prime.junction,
+                    transcript.gene3prime.junction,
                     transcript.transcript1.strand,
                     transcript.transcript2.strand,
                     transcript.transcript1.biotype,

agfusion/parsers.py

@@ -772,6 +772,94 @@ def __init__(self, infile, logger):
 
         self._check_data()
 
+class _CommonFusionFormatBasic(_Parser):
+    """
+    Parent class of CommonFusionFormat* parsers.
+    Defined here: https://github.com/ccmbioinfo/MetaFusion/wiki/metafusion-file-formats#cff-fields
+    """
+
+    def _load_data_indices(self,infile,data_indices):
+        fin = open(infile, "r")
+        n = 0
+        for line in fin.readlines():
+            n += 1
+            line = line.strip().split("\t")
+
+            try:
+                self.fusions.append(
+                    {
+                        "gene5prime": line[data_indices["gene5prime"]],
+                        "gene3prime": line[data_indices["gene3prime"]],
+                        "gene5prime_junction": int(line[data_indices["gene5prime_junction"]]),
+                        "gene3prime_junction": int(line[data_indices["gene3prime_junction"]]),
+                    }
+                )
+            except ValueError as e:
+                print(e)
+                self.logger.warn(
+                    f"Skipping fusion on line {n} because one or more "
+                    + "of the provided breakpoints is not a valid"
+                    + " integer value."
+                )
+
+        fin.close()
+
+        self._check_data()
+
+class CommonFusionFormat(_CommonFusionFormatBasic):
+    """
+    CommonFusionFormat parser.
+    Defined here: https://github.com/ccmbioinfo/MetaFusion/wiki/metafusion-file-formats#cff-fields
+    """
+
+    def __init__(self, infile, logger):
+        super().__init__(logger)
+
+        data_indices = {
+            "gene5prime": 13,
+            "gene3prime": 15,
+            "gene5prime_junction": 1,
+            "gene3prime_junction": 4,
+        }
+
+        self._load_data_indices(infile, data_indices)
+
+class CommonFusionFormatReann(_CommonFusionFormatBasic):
+    """
+    CommonFusionFormat parser.
+    Defined here: https://github.com/ccmbioinfo/MetaFusion/wiki/metafusion-file-formats#cff-fields
+    """
+
+    def __init__(self, infile, logger):
+        super().__init__(logger)
+
+        data_indices = {
+            "gene5prime": 18,
+            "gene3prime": 20,
+            "gene5prime_junction": 1,
+            "gene3prime_junction": 4,
+        }
+
+        self._load_data_indices(infile, data_indices)
+
+class CommonFusionFormatTranscript(_CommonFusionFormatBasic):
+    """
+    CommonFusionFormatTranscript parser.
+    Defined here: https://github.com/ccmbioinfo/MetaFusion/wiki/metafusion-file-formats#cff-fields
+    """
+
+    def __init__(self, infile, logger):
+        super().__init__(logger)
+
+        data_indices = {
+            "gene5prime": 37,
+            "gene3prime": 38,
+            "gene5prime_junction": 1,
+            "gene3prime_junction": 4,
+        }
+
+        self._load_data_indices(infile,data_indices)
+
 
 parsers = {
     "arriba": Arriba,
@@ -791,6 +879,9 @@ def __init__(self, infile, logger):
     "starfusion": STARFusion,
     "tophatfusion": TopHatFusion,
     "mapsplice": MapSplice,
+    "cff": CommonFusionFormat,
+    "cff_reann": CommonFusionFormatReann,
+    "cff_transcript": CommonFusionFormatTranscript,
     # 'fusionseq':FusionSeq,
     # 'prada':Prada,
     # 'gfusion':GFusion,

test/data/FusionsFindingAlgorithms/CommonFusionFormat/fids_out.tsv

@@ -0,0 +1,3 @@
+chr17	35696764	-	chr17	37922746	-	RNA	Sample_A	Tumor	NA	arriba	113	108	ACACA	CDS/splice-site	IKZF3	CDS/splice-site	GeneFusion	ACACA	utr3	IKZF3	cds	True	True	True	True	5.5	24	2154289	-9	F00002817	GTTTGTTTCTTAAAAAAAAAAACCTTAAAATCTTTCTCTCTTTTTCTCTCTAGGTCTTTCTGGAAGTGGATATCTACTCAGACAGTAAGAATTATAAGAG	CCTGGAACAAGTGACAGAAAGGGTTACAAAGGGAACACTGCCAAGGCAGAGAGTTTGTAAATATTTTCTAGAGACCTCAAAAAGGGAGGAGAGTTAAGTT	False	chr17:35696764-35696810	chr17:37922043-37922746	-1	ENST00000353139	ENST00000346872
+chr9	129143450	+	chr9	114476903	-	RNA	Sample_A	Tumor	NA	arriba	1	0	MVB12B	CDS/splice-site	C9orf84	CDS/splice-site	GeneFusion	MVB12B	cds	C9orf84	cds	True	True	True	True	6	291	14531840	-9	F00002851	GACAGCAGATGGTGTGGATGCTGACCTCTGGAAAGACGGCTTATTTAAATCCAAGGTTACCAGATACCTGTGTTTCACAAGATCATTTTCCAAAGAAAAT	CCTATTGAAGCAAAATTAGAAAAAAATTGTCTAAGTATTTGTTTTTTAAGAAACTCCGTGGTAAAAAAGGCTTAAAGAAAATAAGTCAAAGCAGAGACAG	False	chr9:129143343-129143450	chr9:114476725-114476903	-1	ENST00000361171	ENST00000394777
+chr9	114476725	-	chr9	129143343	+	RNA	Sample_A	Tumor	NA	starfusion	85	11	C9orf84	INFRAME	MVB12B	INFRAME	GeneFusion	C9orf84	cds	MVB12B	cds	True	True	True	True	6	291	14531840	-9	F00003651	TGTACAGTTTATTAGGGGGAAAAAGCCTGAAACCAACTACAAGATACAAGAATTGCAATGTCAGATACTAAGTTGGATGCAAAGTCAACAGCAAATTAAG	CCTACACCGGAGAGACAGAGAAAAGGATTGCAGGTAAGGCAGGAAGGAACTGAAGGTGCGCGTGGGCCACTTCTTCCACGCCTGTTCCAATTAACACCAG	False	chr9:114476725-114476903	chr9:129143343-129143450	-1	ENST00000374287	ENST00000361171

test/test_parsers.py

@@ -210,6 +210,95 @@ def test_parse_human(self):
             )
             assert fusion.name in all_fusions, f"{fusion.name} not in list!"
 
+class TestCommonFusionFormat(unittest.TestCase):
+    """Test parse Common Fusion Format"""
+
+    def test_basic(self):
+        """Test basic parsing."""
+
+        all_fusions = ["ACACA_IKZF3", "MVB12B_C9orf84", "C9orf84_MVB12B"]
+        for fusion in parsers.parsers["cff"](
+            f"{BASEDIR}/CommonFusionFormat/" + "fids_out.tsv",
+            db_human.logger,
+        ):
+            fusion = model.Fusion(
+                gene5prime=fusion["gene5prime"],
+                gene5primejunction=fusion["gene5prime_junction"],
+                gene3prime=fusion["gene3prime"],
+                gene3primejunction=fusion["gene3prime_junction"],
+                db=db_human,
+                pyensembl_data=data_human,
+                protein_databases=["pfam"],
+                noncanonical=False,
+            )
+            assert fusion.name in all_fusions, f"{fusion.name} not in list!"
+
+    # def test_with_coding_effect(self):
+    #     """Test parse output with coding effect."""
+
+    #     all_fusions = ["ACACA_IKZF3", "MVB12B_C9orf84", "C9orf84_MVB12B"]
+    #     for fusion in parsers.parsers["cff"](
+    #         f"{BASEDIR}/CommonFusionFormat/" + "fids_out.tsv",
+    #         db_human95.logger,
+    #     ):
+    #         fusion = model.Fusion(
+    #             gene5prime=fusion["gene5prime"],
+    #             gene5primejunction=fusion["gene5prime_junction"],
+    #             gene3prime=fusion["gene3prime"],
+    #             gene3primejunction=fusion["gene3prime_junction"],
+    #             db=db_human95,
+    #             pyensembl_data=data_human95,
+    #             protein_databases=["pfam"],
+    #             noncanonical=False,
+    #         )
+    #         assert fusion.name in all_fusions, f"{fusion.name} not in list!"
+
+class TestCommonFusionFormatReann(unittest.TestCase):
+    """Test parse Common Fusion Format"""
+
+    def test_basic(self):
+        """Test basic parsing."""
+
+        all_fusions = ["ACACA_IKZF3", "MVB12B_C9orf84", "C9orf84_MVB12B"]
+        for fusion in parsers.parsers["cff_reann"](
+            f"{BASEDIR}/CommonFusionFormat/" + "fids_out.tsv",
+            db_human.logger,
+        ):
+            fusion = model.Fusion(
+                gene5prime=fusion["gene5prime"],
+                gene5primejunction=fusion["gene5prime_junction"],
+                gene3prime=fusion["gene3prime"],
+                gene3primejunction=fusion["gene3prime_junction"],
+                db=db_human,
+                pyensembl_data=data_human,
+                protein_databases=["pfam"],
+                noncanonical=False,
+            )
+            assert fusion.name in all_fusions, f"{fusion.name} not in list!"
+
+class TestCommonFusionFormatTranscript(unittest.TestCase):
+    """Test parse Common Fusion Format using Transcript ID columns"""
+
+    def test_basic(self):
+        """Test basic parsing."""
+
+        all_fusions = ["ACACA_IKZF3", "MVB12B_C9orf84", "C9orf84_MVB12B"]
+        for fusion in parsers.parsers["cff_transcript"](
+            f"{BASEDIR}/CommonFusionFormat/" + "fids_out.tsv",
+                db_human.logger,
+        ):
+            fusion = model.Fusion(
+                gene5prime=fusion["gene5prime"],
+                gene5primejunction=fusion["gene5prime_junction"],
+                gene3prime=fusion["gene3prime"],
+                gene3primejunction=fusion["gene3prime_junction"],
+                db=db_human,
+                pyensembl_data=data_human,
+                protein_databases=["pfam"],
+                noncanonical=False,
+            )
+            assert fusion.name in all_fusions, f"{fusion.name} not in list!"
+
 
 if __name__ == "__main__":
     unittest.main()