def test_no_match(self):
""" Example with no FSM match """
conn, cursor = get_db_cursor()
build = "toy_build"
edge_dict = talon.make_edge_dict(cursor)
location_dict = talon.make_location_dict(build, cursor)
run_info = talon.init_run_info(cursor, build)
transcript_dict = talon.make_transcript_dict(cursor, build)
gene_starts, gene_ends = talon.make_gene_start_and_end_dict(cursor, build)
chrom = "chr1"
positions = [1, 100, 500, 600]
strand = "+"
edge_IDs = [2]
vertex_IDs = [2,3,4,5]
v_novelty = [0, 0, 0, 0]
all_matches = talon.search_for_ISM(edge_IDs, transcript_dict)
gene_ID, transcript_ID, novelty, start_end_info = talon.process_FSM(chrom,
positions, strand, edge_IDs,
vertex_IDs, all_matches,
gene_starts, gene_ends,
edge_dict,
location_dict, run_info)
assert gene_ID == transcript_ID == None
conn.close()
def test_FSM_end_diff(self):
""" Example where the transcript is an FSM but has a difference on
the ends large enough to be novel.
"""
conn, cursor = get_db_cursor()
build = "toy_build"
edge_dict = talon.make_edge_dict(cursor)
location_dict = talon.make_location_dict(build, cursor)
run_info = talon.init_run_info(cursor, build)
transcript_dict = talon.make_transcript_dict(cursor, build)
orig_vertices = run_info['vertex']
gene_starts, gene_ends = talon.make_gene_start_and_end_dict(cursor, build)
chrom = "chr2"
positions = [1, 100, 500, 600, 900, 1301] #Last postion is > 300bp away
strand = "+"
edge_IDs = [13, 14, 15]
vertex_IDs = [14, 15, 16, 17]
v_novelty = [0, 0, 0, 0]
all_matches = talon.search_for_ISM(edge_IDs, transcript_dict)
gene_ID, transcript_ID, novelty, start_end_info = talon.process_FSM(chrom,
positions, strand, edge_IDs,
vertex_IDs, all_matches,
gene_starts, gene_ends,
edge_dict,
location_dict, run_info)
correct_gene_ID = fetch_correct_ID("TG2", "gene", cursor)
correct_transcript_ID = fetch_correct_ID("TG2-001", "transcript", cursor)
assert gene_ID == correct_gene_ID
assert transcript_ID == correct_transcript_ID
assert start_end_info["end_vertex"] == orig_vertices + 1
conn.close()
def test_ISM_prefix(self):
""" Example where the transcript is a prefix ISM with a novel start
"""
conn, cursor = get_db_cursor()
build = "toy_build"
edge_dict = talon.make_edge_dict(cursor)
location_dict = talon.make_location_dict(build, cursor)
run_info = talon.init_run_info(cursor, build)
transcript_dict = talon.make_transcript_dict(cursor, build)
gene_starts, gene_ends = talon.make_gene_start_and_end_dict(
cursor, build)
orig_exons = run_info["edge"]
chrom = "chr1"
strand = "+"
positions = [1, 100, 500, 600]
edge_IDs = [2]
vertex_IDs = [2, 3]
v_novelty = [0, 0]
all_matches = talon.search_for_ISM(edge_IDs, transcript_dict)
gene_ID, transcript_ID, novelty, start_end_info = talon.process_ISM(
chrom, positions, strand, edge_IDs, vertex_IDs, all_matches,
transcript_dict, gene_starts, gene_ends, edge_dict, location_dict,
run_info)
correct_gene_ID = fetch_correct_ID("TG1", "gene", cursor)
assert gene_ID == correct_gene_ID
assert start_end_info["vertex_IDs"] == [1, 2, 3, 4]
assert start_end_info["edge_IDs"] == [1, 2, 3]
conn.close()
def test_ISM_suffix(self):
""" Example where the transcript is an ISM with suffix
"""
conn, cursor = get_db_cursor()
build = "toy_build"
edge_dict = talon.make_edge_dict(cursor)
location_dict = talon.make_location_dict(build, cursor)
run_info = talon.init_run_info(cursor, build)
transcript_dict = talon.make_transcript_dict(cursor, build)
gene_starts, gene_ends = talon.make_gene_start_and_end_dict(
cursor, build)
chrom = "chr1"
strand = "+"
positions = [500, 600, 900, 1000]
edge_IDs = [4]
vertex_IDs = [4, 5]
v_novelty = [0, 0]
all_matches = talon.search_for_ISM(edge_IDs, transcript_dict)
gene_ID, transcript_ID, novelty, start_end_info = talon.process_ISM(
chrom, positions, strand, edge_IDs, vertex_IDs, all_matches,
transcript_dict, gene_starts, gene_ends, edge_dict, location_dict,
run_info)
correct_gene_ID = fetch_correct_ID("TG1", "gene", cursor)
assert gene_ID == correct_gene_ID
assert start_end_info["vertex_IDs"] == [3, 4, 5, 6]
assert start_end_info["edge_IDs"] == [3, 4, 5]
assert start_end_info["start_novelty"] == 0 # because the exon is known
assert start_end_info["end_novelty"] == 0
assert transcript_dict[frozenset(start_end_info["edge_IDs"])] != None
conn.close()
def test_find_no_match(self):
""" Example where the toy transcript database contains no matches
for the edge set.
"""
conn, cursor = get_db_cursor()
build = "toy_build"
transcript_dict = talon.make_transcript_dict(cursor, build)
conn.close()
edges = (100, 200, 300)
matches = talon.search_for_ISM(edges, transcript_dict)
# Make sure that no match got returned
assert matches == None
def test_find_monoexon_match(self):
""" Input is a sinlge exon that matches part of an existing transcript
"""
conn, cursor = get_db_cursor()
build = "toy_build"
transcript_dict = talon.make_transcript_dict(cursor, build)
edges = (14, )
matches = talon.search_for_ISM(edges, transcript_dict)
# Make sure that correct match got returned
correct_gene_ID = fetch_correct_ID("TG2", "gene", cursor)
assert matches[0]["gene_ID"] == correct_gene_ID
conn.close()
def test_find_match(self):
""" Example where the toy transcript database contains exactly one
ISM match for the transcript.
"""
conn, cursor = get_db_cursor()
build = "toy_build"
transcript_dict = talon.make_transcript_dict(cursor, build)
edges = (2, 3)
matches = talon.search_for_ISM(edges, transcript_dict)
# Make sure that correct match got returned
correct_gene_ID = fetch_correct_ID("TG1", "gene", cursor)
assert matches[0]["gene_ID"] == correct_gene_ID
conn.close()
def test_no_match(self):
""" Example with no ISM match """
conn, cursor = get_db_cursor()
build = "toy_build"
edge_dict = talon.make_edge_dict(cursor)
location_dict = talon.make_location_dict(build, cursor)
run_info = talon.init_run_info(cursor, build)
transcript_dict = talon.make_transcript_dict(cursor, build)
gene_starts, gene_ends = talon.make_gene_start_and_end_dict(
cursor, build)
chrom = "chr1"
strand = "+"
positions = [1, 100, 900, 1000]
edge_IDs = [200]
vertex_IDs = [2, 5]
v_novelty = [0, 0]
all_matches = talon.search_for_ISM(edge_IDs, transcript_dict)
assert all_matches == None
conn.close()
def test_FSM_perfect(self):
""" Example where the transcript is a perfect full splice match.
"""
conn, cursor = get_db_cursor()
build = "toy_build"
edge_dict = talon.make_edge_dict(cursor)
location_dict = talon.make_location_dict(build, cursor)
run_info = talon.init_run_info(cursor, build)
transcript_dict = talon.make_transcript_dict(cursor, build)
gene_starts, gene_ends = talon.make_gene_start_and_end_dict(cursor, build)
chrom = "chr1"
positions = [1, 100, 500, 600, 900, 1010]
strand = "+"
edge_IDs = [2, 3, 4]
vertex_IDs = [2, 3, 4, 5]
v_novelty = [0, 0, 0, 0]
all_matches = talon.search_for_ISM(edge_IDs, transcript_dict)
gene_ID, transcript_ID, novelty, start_end_info = talon.process_FSM(chrom,
positions, strand, edge_IDs,
vertex_IDs, all_matches,
gene_starts, gene_ends,
edge_dict,
location_dict, run_info)
correct_gene_ID = fetch_correct_ID("TG1", "gene", cursor)
correct_transcript_ID = fetch_correct_ID("TG1-001", "transcript", cursor)
assert gene_ID == correct_gene_ID
assert transcript_ID == correct_transcript_ID
assert novelty == []
assert start_end_info["start_vertex"] == 1
assert start_end_info["end_vertex"] == 6
assert start_end_info["diff_3p"] == 10
conn.close()