def test_store_promoters(self):
promoters = [
Promoter("gene1", 10, 20, seq=Seq("cgtacgtacgt")),
Promoter("gene2", 30, 40, seq=Seq("cgtacgtacgt")),
CombinedPromoter("gene3", "gene4", 50, 60, seq=Seq("cgtacgtacgt"))
]
record_with_promoters = create_fake_record()
cassis.store_promoters(
promoters,
record_with_promoters) # add ("store") promoters to seq_record
record_without_promoters = create_fake_record(
) # just the same, without adding promoters
# test promoter features
expected_count = record_without_promoters.get_feature_count() + len(
promoters)
assert expected_count == record_with_promoters.get_feature_count()
for i in range(len(promoters)):
feature = record_with_promoters.get_generics()[i]
assert feature.type == "promoter"
assert feature.get_qualifier("seq") == ("cgtacgtacgt", )
# especially test bidirectional promoter feature (third promoter, last feature)
last_promoter = record_with_promoters.get_generics()[-1]
assert last_promoter.get_qualifier("locus_tag") == ("gene3", "gene4")
assert last_promoter.notes == ["bidirectional promoter"]
def test_regeneration(self):
record = create_fake_record()
results = cassis.CassisResults(record.id)
# create a prediction, since it will generate a border with many extra qualifiers
start_marker = ClusterMarker("gene1", Motif(3, 3, score=1))
start_marker.promoter = "gene1"
start_marker.abundance = 2
end_marker = ClusterMarker("gene4", Motif(3, 3, score=1))
end_marker.promoter = "gene3+gene4"
assert end_marker.abundance == 1
cluster = cassis.ClusterPrediction(start_marker, end_marker)
results.subregions = cassis.create_subregions("gene1", [cluster],
record)
assert results.subregions
results.promoters = [
Promoter("gene1", 10, 20, seq=Seq("cgtacgtacgt")),
Promoter("gene2", 30, 40, seq=Seq("cgtacgtacgt")),
CombinedPromoter("gene3", "gene4", 50, 60, seq=Seq("cgtacgtacgt"))
]
round_trip = cassis.regenerate_previous_results(
results.to_json(), record, None)
assert isinstance(round_trip, cassis.CassisResults)
assert len(results.subregions) == len(round_trip.subregions)
for old, new in zip(results.subregions, round_trip.subregions):
assert old.location == new.location
assert old.to_biopython()[0].qualifiers == new.to_biopython(
)[0].qualifiers
assert round_trip.promoters == results.promoters
def test_check_cluster_predictions(self):
seq_record = create_fake_record()
promoters = [
Promoter("gene1", 1, 5),
Promoter("gene2", 6, 10),
CombinedPromoter("gene3", "gene4", 11, 15)
]
ignored_genes = [ # see captured logging
Gene(FeatureLocation(1, 5), locus_tag="gene5")
]
clusters = [
ClusterPrediction(ClusterMarker("gene1", Motif(3, 3, score=1)),
ClusterMarker("gene4", Motif(3, 3, score=1)))
]
expected = [
ClusterPrediction(ClusterMarker("gene1", Motif(3, 3, score=1)),
ClusterMarker("gene4", Motif(3, 3, score=1)))
]
expected[0].start.promoter = "gene1"
expected[0].end.promoter = "gene3+gene4"
expected[0].genes = 4
expected[0].promoters = 3
assert check_cluster_predictions(clusters, seq_record, promoters,
ignored_genes) == expected
def test_get_anchor_promoter(self):
anchor = "gene3"
promoters = [
Promoter("gene1", 1, 1),
Promoter("gene2", 2, 2),
CombinedPromoter("gene3", "gene4", 3, 4),
Promoter("gene5", 5, 5)
]
self.assertEqual(get_anchor_promoter_index(anchor, promoters), 2)
def test_filter_fimo_results(self):
fimo_dir = os.path.join(self.options.output_dir, "fimo")
motifs = [Motif(0, 3)]
# gene2 will be the anchor promoter
anchor_promoter = 1
promoters = []
for i in range(1, 16):
promoters.append(Promoter("gene%d" % i, i * 10, i * 10 + 4))
# need certain amount of promoters, otherwise the proportion of
# promoters with a motif (motif frequency) will be too high --> error
expected_motifs = [Motif(0, 3, hits={"gene1": 1, "gene2": 2})]
# fake FIMO output file, corresponding to expected_motifs
source = path.get_full_path(__file__, "data", "fake_short_fimo.txt")
target = os.path.join(fimo_dir, "+00_-03")
if not os.path.exists(target):
os.makedirs(target)
copy(source, os.path.join(target,
"fimo.txt")) # overwrite fimo.txt if exists
found_motifs = filter_fimo_results(motifs, fimo_dir, promoters,
anchor_promoter)
assert found_motifs == expected_motifs
bs_per_promoter, expected_bs_per_promoter = read_generated_expected_file(
os.path.join(target, "bs_per_promoter.csv"),
"expected_bs_per_promoter.csv")
self.assertEqual(bs_per_promoter, expected_bs_per_promoter)
def test_get_promoter_sets(self):
meme_dir = os.path.join(self.options.output_dir, "meme")
anchor_promoter = 5
promoters = [
Promoter("gene1", 1, 1, seq=Seq("acgtacgtacgtacgt")),
Promoter("gene2", 2, 2, seq=Seq("acgtacgtacgtacgt")),
CombinedPromoter("gene3",
"gene4",
3,
4,
seq=Seq("acgtacgtacgtacgt")),
Promoter("gene5", 5, 5, seq=Seq("acgtacgtacgtacgt")),
Promoter("gene6", 6, 6, seq=Seq("acgtacgtacgtacgt")),
# promoter with index=5 --> anchor promoter
Promoter("gene7", 7, 7, seq=Seq("acgtacgtacgtacgt")),
Promoter("gene8", 8, 8, seq=Seq("acgtacgtacgtacgt")),
Promoter("gene9", 9, 9, seq=Seq("acgtacgtacgtacgt"))
]
expected_motifs = [
Motif(plus, minus) for plus in range(3)
for minus in range(3 - plus, 6)
]
self.assertEqual(generate_motifs(meme_dir, anchor_promoter, promoters),
expected_motifs)
def test_serialisation(self):
for seq in [Seq("ACGT"), "ACGT"]:
for cls, promoter in [(Promoter, Promoter("gene1", 1, 5, seq=seq)),
(CombinedPromoter,
CombinedPromoter("gene1",
"gene2",
2,
7,
seq=seq))]:
round_trip = cls.from_json(promoter.to_json())
assert promoter.seq == round_trip.seq
assert round_trip == promoter
def test_get_islands(self):
motifs = [Motif(0, 3, hits={"gene1": 1, "gene2": 2}),
Motif(0, 4, hits={"gene2": 3, "gene4": 2, "gene5": 1})]
# gene2 will be the anchor promoter
anchor_promoter = 1
promoters = []
for i in range(1, 7):
promoters.append(Promoter("gene%d" % i, i * 10, i * 10 + 4))
# resulting in 2 different islands (this example)
# promoter (pos): 1 2 3 4 5 6
# binding sites: 1 2 0 0 0 0
# island: |-|
first_island = Island(promoters[0], promoters[1], motifs[0])
# promoter (pos): 1 2 3 4 5 6
# binding sites: 0 3 0 2 1 0
# island: |---|
second_island = Island(promoters[1], promoters[4], motifs[1])
expected_islands = [first_island, second_island]
assert get_islands(anchor_promoter, motifs, promoters) == expected_islands
def test_sort_by_abundance(self):
islands = []
# island 1: [gene1 -- gene2]
motif = Motif(0, 3, score=3, hits={"gene1": 1, "gene2": 1})
islands.append(Island(Promoter("gene1", 1, 1), Promoter("gene2", 2, 2), motif))
# island 2: [gene2 -- gene5]
motif = Motif(3, 0, score=2, hits={"gene2": 1, "gene3": 1, "gene4": 1, "gene5": 1})
islands.append(Island(Promoter("gene2", 2, 2), Promoter("gene5", 5, 5), motif))
# island 3: [gene1 -- gene5]
motif = Motif(3, 3, score=1, hits={"gene1": 1, "gene2": 1, "gene3": 1, "gene4": 1, "gene5": 1})
islands.append(Island(Promoter("gene1", 1, 1), Promoter("gene5", 5, 5), motif))
# left border: 2x gene1, 1x gene2
# right border: 2x gene5, 1x gene2
expected_clusters = []
# cluster 1: [gene1 -- gene5] --> abundance 2+2 (most abundant)
start = ClusterMarker("gene1", Motif(3, 3, score=1))
start.abundance = 2
end = ClusterMarker("gene5", Motif(3, 3, score=1))
end.abundance = 2
expected_clusters.append(ClusterPrediction(start, end))
# cluster 3: [gene2 -- gene5] --> abundance 1+2, score 2+1 (better/lower)
start = ClusterMarker("gene2", Motif(3, 0, score=2))
start.abundance = 1
end = ClusterMarker("gene5", Motif(3, 3, score=1))
end.abundance = 2
expected_clusters.append(ClusterPrediction(start, end))
# cluster 2: [gene1 -- gene2] --> abundance 2+1, score 1+3 (worse, higher)
start = ClusterMarker("gene1", Motif(3, 3, score=1))
start.abundance = 2
end = ClusterMarker("gene2", Motif(0, 3, score=3))
end.abundance = 1
expected_clusters.append(ClusterPrediction(start, end))
# cluster 4: [gene2 -- gene2] --> abundance 1+1
start = ClusterMarker("gene2", Motif(3, 0, score=2))
start.abundance = 1
end = ClusterMarker("gene2", Motif(0, 3, score=3))
end.abundance = 1
expected_clusters.append(ClusterPrediction(start, end))
# abundance: as high as possible
# score: as low as possible
self.assertEqual(create_predictions(islands), expected_clusters)
def test_promoter_id(self):
assert Promoter("gene1", 1, 5).get_id() == "gene1"
assert CombinedPromoter("gene1", "gene2", 1,
5).get_id() == "gene1+gene2"
def setUp(self):
self.promoter = Promoter("gene1", 1, 1)