Merge pull request murphycj#50 from murphycj/bug/fix-fusioninspector-…

…parser

Fix Fusioninspector parser

.github/workflows/test.yaml

@@ -11,7 +11,7 @@ jobs:
         python-version: ["3.7", "3.8", "3.9", "3.10"]
     steps:
       - name: Checkout source code
-        uses: actions/checkout@v1
+        uses: actions/checkout@v3
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip

agfusion/parsers.py

@@ -701,30 +701,22 @@ class FusionInspector(_Parser):
     def __init__(self, infile, logger):
         super().__init__(logger)
 
-        fin = open(infile, "r")
-        for line in fin.readlines():
-            if re.findall(r"^#", line):
-                line = line.rstrip().split("\t")
-                if line[0] != "#FusionName" and line[0] != "#fusion_name":
-                    raise AssertionError(
-                        "Unrecognized FusionInspector input for first column"
-                        + " in header. Should be #FusionName or #fusion_name."
-                    )
+        data = pd.read_csv(infile, delimiter="\t")
+        data.columns = [i.replace("#", "") for i in data.columns]
 
-                assert line[3] == "LeftGene", "Unrecognized " + "FusionInspector input"
-                assert line[5] == "LeftBreakpoint", "Unrecognized " + "FusionInspector input"
-                assert line[6] == "RightGene", "Unrecognized " + "FusionInspector input"
-                assert line[8] == "RightBreakpoint", "Unrecognized " + "FusionInspector input"
-                continue
+        cols = ["LeftGene", "LeftBreakpoint", "RightGene", "RightBreakpoint"]
+        assert all(
+            i in data.columns for i in cols
+        ), "Unrecognized FusionInspector input. Could not find all columns: " + ",".join(cols)
 
-            line = line.strip().split("\t")
+        for i in data.index:
 
-            gene_5prime = line[3].split("^")[1].split(".")[0]
-            gene_5prime_name = line[3].split("^")[0]
-            gene_5prime_junction = int(line[5].split(":")[1])
-            gene_3prime = line[6].split("^")[1].split(".")[0]
-            gene_3prime_name = line[6].split("^")[0]
-            gene_3prime_junction = int(line[8].split(":")[1])
+            gene_5prime = data.at[i, "LeftGene"].split("^")[1].split(".")[0]
+            gene_5prime_name = data.at[i, "LeftGene"].split("^")[0]
+            gene_5prime_junction = int(data.at[i, "LeftBreakpoint"].split(":")[1])
+            gene_3prime = data.at[i, "RightGene"].split("^")[1].split(".")[0]
+            gene_3prime_name = data.at[i, "RightGene"].split("^")[0]
+            gene_3prime_junction = int(data.at[i, "RightBreakpoint"].split(":")[1])
             self.fusions.append(
                 {
                     "gene5prime": gene_5prime,
@@ -735,7 +727,6 @@ def __init__(self, infile, logger):
                     "gene3prime_junction": gene_3prime_junction,
                 }
             )
-        fin.close()
 
         self._check_data()
 

test/data/FusionsFindingAlgorithms/FusionInspector/test.FusionInspector.fusions.abridged.txt

@@ -0,0 +1,3 @@
+#FusionName	JunctionReadCount	SpanningFragCount	est_J	est_S	LeftGene	LeftLocalBreakpoint	LeftBreakpoint	RightGene	RightLocalBreakpoint	RightBreakpoint	SpliceType	LargeAnchorSupport	NumCounterFusionLeft	NumCounterFusionRight	FAR_left	FAR_right	LeftBreakDinuc	LeftBreakEntropy	RightBreakDinuc	RightBreakEntropy	FFPM	microh_brkpt_dist	num_microh_near_brkpt
+AL627171.2--TPM3	1551	3	1538.54	3.00	AL627171.2^ENSG00000282885.2	2641	chr14:49862686:-	TPM3^ENSG00000143549.21	23194	chr1:154166382:-	INCL_NON_REF_SPLICE	NO	152	89	10.16	17.28	GT	1.5058	AG	1.8892	26.6109	1	30
+STAT3--AL627171.2	955	1	934.46	1.00	STAT3^ENSG00000168610.16	21538	chr17:42321234:-	AL627171.2^ENSG00000282885.2	32742	chr14:49862799:-	INCL_NON_REF_SPLICE	YES	0	152	957.00	6.25	GT	1.8295	AG	1.7968	16.1485	1	25

test/test_parsers.py

@@ -20,6 +20,9 @@
 db_human95.build = "homo_sapiens_95"
 
 
+BASEDIR = "./data/FusionsFindingAlgorithms"
+
+
 class TestFusionCatcher(unittest.TestCase):
     """Test parse FusionCatcher parse."""
 
@@ -37,7 +40,7 @@ def test_parse(self):
             "C920009B18Rik_H60b",
         ]
         for fusion in parsers.parsers["fusioncatcher"](
-            "./data/FusionsFindingAlgorithms/FusionCatcher/final-list_candidate-fusion-genes.txt",
+            f"{BASEDIR}/FusionCatcher/final-list_candidate-fusion-genes.txt",
             db_mouse.logger,
         ):
             fusion = model.Fusion(
@@ -63,7 +66,7 @@ def test_parse(self):
             "BCR_ABL1",
         ]
         for fusion in parsers.parsers["arriba"](
-            "./data/FusionsFindingAlgorithms/Arriba/fusions.tsv",
+            f"{BASEDIR}/Arriba/fusions.tsv",
             db_human.logger,
         ):
             fusion = model.Fusion(
@@ -87,8 +90,7 @@ def test_basic(self):
 
         all_fusions = ["ACACA_STAC2", "RPS6KB1_SNF8"]
         for fusion in parsers.parsers["starfusion"](
-            "./data/FusionsFindingAlgorithms/STARFusion/"
-            + "star-fusion.fusion_candidates.final.abridged",
+            f"{BASEDIR}/STARFusion/" + "star-fusion.fusion_candidates.final.abridged",
             db_human.logger,
         ):
             fusion = model.Fusion(
@@ -108,8 +110,7 @@ def test_with_coding_effect(self):
 
         all_fusions = ["ARID3B_MYCNUT", "ARID3B_MYCN", "TVP23C_CDRT4"]
         for fusion in parsers.parsers["starfusion"](
-            "./data/FusionsFindingAlgorithms/STARFusion/"
-            + "star-fusion.fusion_predictions.abridged.coding_effect.tsv",
+            f"{BASEDIR}/STARFusion/" + "star-fusion.fusion_predictions.abridged.coding_effect.tsv",
             db_human95.logger,
         ):
             fusion = model.Fusion(
@@ -133,7 +134,7 @@ def test_parse_mouse(self):
 
         all_fusions = ["Mocos_Rprd1a", "Ubc_Ubb", "Ubc_Gm11808", "Gm21887_Gm47283"]
         for fusion in parsers.parsers["longgf"](
-            "./data/FusionsFindingAlgorithms/LongGF/fusions_mouse.log",
+            f"{BASEDIR}/LongGF/fusions_mouse.log",
             db_mouse.logger,
         ):
             fusion = model.Fusion(
@@ -153,7 +154,48 @@ def test_parse_human(self):
 
         all_fusions = ["BCAS4_BCAS3", "HNRNPC_ACIN1"]
         for fusion in parsers.parsers["longgf"](
-            "./data/FusionsFindingAlgorithms/LongGF/fusions_hg38.log",
+            f"{BASEDIR}/LongGF/fusions_hg38.log",
+            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 TestFusionInspector(unittest.TestCase):
+    """Test parse FusionInspector"""
+
+    def test_parse_human(self):
+        """Test basic parsing."""
+
+        all_fusions = ["AL627171.2_TPM3", "STAT3_AL627171.2"]
+
+        for fusion in parsers.parsers["fusioninspector"](
+            f"{BASEDIR}/FusionInspector/test.FusionInspector.fusions.abridged.txt",
+            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!"
+
+        for fusion in parsers.parsers["fusioninspector"](
+            f"{BASEDIR}/FusionInspector/test.FusionInspector.fusions.txt",
             db_human95.logger,
         ):
             fusion = model.Fusion(