Rename GeneMatrixTransposed to GeneSets, add minimal tests

README.md

@@ -12,9 +12,9 @@ Pandas API for Gene Set Enrichment Analysis in Python (GSEApy, cudaGSEA, GSEA)
 from pandas import read_table
 from gsea_api.expression_set import ExpressionSet
 from gsea_api.gsea import GSEADesktop
-from gsea_api.molecular_signatures_db import GeneMatrixTransposed
+from gsea_api.molecular_signatures_db import GeneSets
 
-reactome_pathways = GeneMatrixTransposed.from_gmt('ReactomePathways.gmt')
+reactome_pathways = GeneSets.from_gmt('ReactomePathways.gmt')
 
 gsea = GSEADesktop()
 

gsea_api/gsea/java.py

@@ -7,7 +7,7 @@
 from IPython.core.display import display
 from pandas import read_table, DataFrame
 
-from ..molecular_signatures_db import GeneMatrixTransposed
+from ..molecular_signatures_db import GeneSets
 from .base import GSEA
 from ..paths import tmp_dir, third_party_dir
 

gsea_api/molecular_signatures_db.py

@@ -16,7 +16,7 @@ def from_gmt_line(cls, line):
         return cls(name, ids)
 
 
-class GeneMatrixTransposed:
+class GeneSets:
 
     def __init__(self, gene_sets, name=''):
         self.gene_sets = gene_sets
@@ -31,7 +31,7 @@ def from_gmt(cls, path, name=''):
             }, name=name)
 
     def trim(self, min_genes, max_genes: int):
-        return GeneMatrixTransposed({
+        return GeneSets({
             gene_set
             for gene_set in self.gene_sets
             if min_genes <= len(gene_set.genes) <= max_genes
@@ -54,7 +54,7 @@ def to_gmt(self, path):
             self._to_gmt(path)
 
     def subset(self, genes: Set[str]):
-        return GeneMatrixTransposed({
+        return GeneSets({
             GeneSet(name=gene_set.name, genes=gene_set.genes & genes)
             for gene_set in self.gene_sets
         })
@@ -68,6 +68,13 @@ def all_identifiers(self):
 
         return all_identifiers
 
+    def __len__(self):
+        return len(self.gene_sets)
+
+
+# for backwards compatibility
+GeneMatrixTransposed = GeneSets
+
 
 class MolecularSignaturesDatabase:
     def __init__(self, path, version='6.2'):
@@ -90,7 +97,7 @@ def resolve(self, gene_sets, id_type):
         else:
             raise ValueError(f'Unknown library: {path}!')
 
-    def load(self, gene_sets, id_type) -> GeneMatrixTransposed:
+    def load(self, gene_sets, id_type) -> GeneSets:
         path = self.resolve(gene_sets=gene_sets, id_type=id_type)
 
-        return GeneMatrixTransposed.from_gmt(path, name=gene_sets)
+        return GeneSets.from_gmt(path, name=gene_sets)

tests/README.md

@@ -0,0 +1,3 @@
+## Legal
+
+A subset of Gene Ontology gene sets is used in tests. Gene Ontology is licenced under [Creative Commons Attribution 4.0 Unported License](https://creativecommons.org/licenses/by/4.0/legalcode).

tests/gene_ontology_pathways.gmt

@@ -0,0 +1,4 @@
+GO:0061038	uterus morphogenesis	ASH1L	KDM5B	STRA6	WNT7A	WNT9B	NIPBL
+GO:0048265	response to pain	TAC1	COMT	TSPO	SLC6A2	TRPA1	P2RX3	THBS4	TACR1	DBH	PRKCG	GCH1	NMUR2	P2RX4	EDNRB	THBS1	CACNA1A	CRH	CACNA1B	CAPN2	UCN	RET	SCN9A	VWA1	LPAR5	GJA4	P2RX2	RELN	TRPV1	NTRK1	PIRT
+GO:0061366	behavioral response to chemical pain	P2RX3	NTRK1
+GO:0061368	behavioral response to formalin induced pain	P2RX3	NTRK1

tests/test_gene_sets.py

@@ -0,0 +1,9 @@
+from gsea_api.molecular_signatures_db import GeneSets
+
+
+def test_from_gmt():
+    pathways = GeneSets.from_gmt('tests/gene_ontology_pathways.gmt')
+    assert len(pathways) == len(pathways.gene_sets)
+    assert len(pathways.gene_sets) == 4
+
+    assert len(pathways.trim(min_genes=3, max_genes=100)) == 2