def test_plot_genome_coverage_minimap():
""" Replace BLAST results with minimap2. """
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage
1-HCV-1a,HCV-1a,1,8001,0,0,5
1-HCV-1a,HCV-1a,2,8002,0,0,5
1-HCV-1a,HCV-1a,3,8003,0,0,7
1-HCV-1a,HCV-1a,4,8004,0,0,5
1-HCV-1a,HCV-1a,5,8005,0,0,5
1-HCV-1a,HCV-1a,6,8006,0,0,5
""")
minimap_hits_csv = StringIO("""\
contig,ref_name,score,match,pident,start,end,ref_start,ref_end
1-HCV-1a,HCV-1a,40,0.33,100,5,6,7006,7005
1-HCV-1a,HCV-1a,50,0.5,100,1,3,8001,8003
""")
expected_figure = """\
5'[1-341], C[342-914], E1[915-1490], E2[1491-2579], p7[2580-2768], \
NS2[2769-3419], NS3[3420-5312], NS4b[5475-6257], NS4a[5313-5474], \
NS5a[6258-7601], NS5b[7602-9374], 3'[9375-9646]
7005<-1.2--7006, 8001--1.1->8003
8001--1.1->8003, 8005--1.2->8006
Coverage 5x2, 7, 5x3
[8001-8006], 1-HCV-1a - depth 7(1-9646)
"""
figure = build_coverage_figure(genome_coverage_csv, minimap_hits_csv)
assert summarize_figure(figure) == expected_figure
def test_plot_genome_coverage_blast_collision():
""" Two blast results end at the same position. """
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage
1-HCV-1a,HCV-1a,1,8001,0,0,5
1-HCV-1a,HCV-1a,2,8002,0,0,5
1-HCV-1a,HCV-1a,3,8003,0,0,7
1-HCV-1a,HCV-1a,4,8004,0,0,5
1-HCV-1a,HCV-1a,5,8005,0,0,5
1-HCV-1a,HCV-1a,6,8006,0,0,5
""")
blast_csv = StringIO("""\
contig_num,ref_name,score,match,pident,start,end,ref_start,ref_end
1,HCV-1g,30,0.33,90,1,2,5001,5002
1,HCV-1a,40,0.33,100,3,6,7003,7006
1,HCV-1a,50,0.5,100,1,6,8001,8006
""")
expected_figure = """\
5'[1-341], C[342-914], E1[915-1490], E2[1491-2579], p7[2580-2768], \
NS2[2769-3419], NS3[3420-5312], NS4b[5475-6257], NS4a[5313-5474], \
NS5a[6258-7601], NS5b[7602-9374], 3'[9375-9646]
7003--1.2->7006, 8001--1.1->8006
8001--1.1->8006, 8003--1.2->8006
Coverage 5x2, 7, 5x3
[8001-8006], 1-HCV-1a - depth 7(1-9646)
"""
figure = build_coverage_figure(genome_coverage_csv, blast_csv)
assert summarize_figure(figure) == expected_figure
def test_plot_genome_coverage_blast_aligns_refs():
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,1,2261,0,0,5
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,2,2262,0,0,5
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,3,2263,0,0,5
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,4,2264,0,0,5
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,5,2265,0,0,5
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,6,2266,0,0,5
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,7,2267,0,0,5
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,8,2268,0,0,5
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,9,2269,0,0,5
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,10,2270,0,0,5
""")
blast_csv = StringIO("""\
contig_num,ref_name,score,match,pident,start,end,ref_start,ref_end
1,HIV1-G-CM-KP718923-seed,300,1,90,1,10,1653,1662
""")
expected_figure = """\
5' LTR[1-634], gag[789-2289], vif[5040-5616], tat[8379-8469], nef[8796-9414]
tat[5831-6045], vpu[6061-6307], rev[8379-8653], 3' LTR[9086-9719]
pol[2085-5096], vpr[5558-5847], rev[5970-6045], env[6225-8795]
PR[2252-2549], RT[2549-3869], INT[4229-5093], V3[7109-7217], GP41[7757-8792]
2261--1.1->2270
2261--1.1->2270
Coverage 5x10
[2261-2270], 1-HIV1-G-CM-KP718923-seed - depth 5(1-9719)
"""
figure = build_coverage_figure(genome_coverage_csv, blast_csv)
assert summarize_figure(figure) == expected_figure
def test_plot_genome_coverage_unmapped():
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage,link
1-HCV-1a,HCV-1a,1,1,0,0,5,M
1-HCV-1a,HCV-1a,2,2,0,0,5,M
1-HCV-1a,HCV-1a,3,3,0,0,5,M
1-HCV-1a,HCV-1a,4,4,0,0,6,M
1-HCV-1a,HCV-1a,5,5,0,0,6,U
1-HCV-1a,HCV-1a,6,6,0,0,6,U
1-HCV-1a,HCV-1a,7,7,0,0,7,U
1-HCV-1a,HCV-1a,8,8,0,0,7,U
1-HCV-1a,HCV-1a,9,9,0,0,7,U
1-HCV-1a,HCV-1a,10,10,0,0,8,M
1-HCV-1a,HCV-1a,11,11,0,0,8,M
1-HCV-1a,HCV-1a,12,12,0,0,8,M
""")
expected_figure = """\
5'[1-341], C[342-914], E1[915-1490], E2[1491-2579], p7[2580-2768], \
NS2[2769-3419], NS3[3420-5312], NS4b[5475-6257], NS4a[5313-5474], \
NS5a[6258-7601], NS5b[7602-9374], 3'[9375-9646]
Coverage 5x3, 6x3, 7x3, 8x3
[1-12], 1-HCV-1a - depth 8(1-9646), yellow{5-9}
"""
figure = build_coverage_figure(genome_coverage_csv)
assert summarize_figure(figure) == expected_figure
def test_plot_genome_coverage_gap():
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage
1-HCV-1a,HCV-1a,1,1,0,0,5
1-HCV-1a,HCV-1a,2,2,0,0,5
1-HCV-1a,HCV-1a,4,4,0,0,6
1-HCV-1a,HCV-1a,5,5,0,0,6
1-HCV-1a,HCV-1a,6,6,0,0,6
contig-1-HCV-1a,HCV-1a,1,1,0,,
contig-1-HCV-1a,HCV-1a,2,2,0,,
contig-1-HCV-1a,HCV-1a,4,4,0,,
contig-1-HCV-1a,HCV-1a,5,5,0,,
contig-1-HCV-1a,HCV-1a,6,6,0,,
""")
expected_figure = """\
5'[1-341], C[342-914], E1[915-1490], E2[1491-2579], p7[2580-2768], \
NS2[2769-3419], NS3[3420-5312], NS4b[5475-6257], NS4a[5313-5474], \
NS5a[6258-7601], NS5b[7602-9374], 3'[9375-9646]
Coverage 5x2, 0, 6x3
[1-2], [4-6], 1-HCV-1a - depth 6(1-9646)
[1-2], [4-6], contig-1-HCV-1a(1-9646)
"""
figure = build_coverage_figure(genome_coverage_csv)
assert expected_figure == summarize_figure(figure)
def test_plot_genome_coverage_insertion():
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage
1-HCV-1a,HCV-1a,1,1,0,0,5
1-HCV-1a,HCV-1a,2,2,0,0,5
1-HCV-1a,HCV-1a,3,3,0,0,5
1-HCV-1a,HCV-1a,4,,0,0,6
1-HCV-1a,HCV-1a,5,,0,0,6
1-HCV-1a,HCV-1a,6,,0,0,6
1-HCV-1a,HCV-1a,7,4,0,0,7
1-HCV-1a,HCV-1a,8,5,0,0,7
1-HCV-1a,HCV-1a,9,6,0,0,7
1-HCV-1a,HCV-1a,10,7,0,0,8
1-HCV-1a,HCV-1a,11,8,0,0,8
1-HCV-1a,HCV-1a,12,9,0,0,8
""")
expected_figure = """\
5'[1-341], C[342-914], E1[915-1490], E2[1491-2579], p7[2580-2768], \
NS2[2769-3419], NS3[3420-5312], NS4b[5475-6257], NS4a[5313-5474], \
NS5a[6258-7601], NS5b[7602-9374], 3'[9375-9646]
Coverage 5x3, 7x3, 8x3
[1-9], 1-HCV-1a - depth 8(1-9646), lightgreen{4-6}
"""
figure = build_coverage_figure(genome_coverage_csv)
assert expected_figure == summarize_figure(figure)
def test_plot_genome_coverage_two_contigs():
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,dels,coverage
1-HCV-1a,HCV-1a,1,1,0,5
1-HCV-1a,HCV-1a,2,2,0,5
1-HCV-1a,HCV-1a,3,3,0,7
1-HCV-1a,HCV-1a,4,4,0,5
1-HCV-1a,HCV-1a,5,5,0,5
1-HCV-1a,HCV-1a,6,6,0,5
contig-1-HCV-1a,HCV-1a,1,1,,
contig-1-HCV-1a,HCV-1a,2,2,,
contig-1-HCV-1a,HCV-1a,3,3,,
contig-1-HCV-1a,HCV-1a,4,4,,
contig-1-HCV-1a,HCV-1a,5,5,,
contig-1-HCV-1a,HCV-1a,6,6,,
2-HCV-1b-partial,,1,,0,5
2-HCV-1b-partial,,2,,0,5
2-HCV-1b-partial,,3,,0,29
2-HCV-1b-partial,,4,,0,5
2-HCV-1b-partial,,5,,0,5
2-HCV-1b-partial,,6,,0,5
3-HCV-1a,HCV-1a,101,101,0,15
3-HCV-1a,HCV-1a,102,102,0,15
3-HCV-1a,HCV-1a,103,103,0,17
3-HCV-1a,HCV-1a,104,104,0,15
3-HCV-1a,HCV-1a,105,105,0,15
3-HCV-1a,HCV-1a,106,106,0,15
contig-3-HCV-1a,HCV-1a,1,101,,
contig-3-HCV-1a,HCV-1a,2,102,,
contig-3-HCV-1a,HCV-1a,3,103,,
contig-3-HCV-1a,HCV-1a,4,104,,
contig-3-HCV-1a,HCV-1a,5,105,,
contig-3-HCV-1a,HCV-1a,6,106,,
""")
expected_figure = """\
5'[1-341], C[342-914], E1[915-1490], E2[1491-2579], p7[2580-2768], \
NS2[2769-3419], NS3[3420-5312], NS4b[5475-6257], NS4a[5313-5474], \
NS5a[6258-7601], NS5b[7602-9374], 3'[9375-9646]
Coverage 15x2, 17, 15x3
[101-106], 3-HCV-1a - depth 17(1-9646)
[101-106], contig-3-HCV-1a(1-9646)
Coverage 5x2, 7, 5x3
[1-6], 1-HCV-1a - depth 7(1-9646)
[1-6], contig-1-HCV-1a(1-9646)
[1-500], [1001-1500], [2001-2500], [3001-3500], [4001-4500], \
[5001-5500], [6001-6500], [7001-7500], [8001-8500], [9001-9500], \
Partial Blast Results(1-9646)
Coverage 5x2, 29, 5x3
[1-6], 2-HCV-1b-partial - depth 29(1-9646)
"""
figure = build_coverage_figure(genome_coverage_csv)
assert summarize_figure(figure) == expected_figure
def test_plot_genome_coverage_empty():
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage
""")
expected_figure = """\
No contigs found.(1-500)
"""
figure = build_coverage_figure(genome_coverage_csv)
assert expected_figure == summarize_figure(figure)
def test_plot_genome_coverage_zero():
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,dels,coverage
1-HCV-1b-partial,,1,,0,0
1-HCV-1b-partial,,2,,0,0
1-HCV-1b-partial,,3,,0,0
1-HCV-1b-partial,,4,,0,0
1-HCV-1b-partial,,5,,0,0
1-HCV-1b-partial,,6,,0,0
""")
expected_figure = """\
[1-500], Partial Blast Results(1-500)
[1-6], 1-HCV-1b-partial(1-500)
"""
figure = build_coverage_figure(genome_coverage_csv)
assert expected_figure == summarize_figure(figure)
def test_plot_genome_coverage_partial_header():
""" Last dash in the header banner can be less than 500 wide. """
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage
""")
genome_coverage_csv.seek(0, 2) # EOF
for i in range(1010):
genome_coverage_csv.write(f'1-HCV-1a-partial,,{i+1},,0,0,5\n')
genome_coverage_csv.seek(0)
expected_figure = """\
[1-500], [1001-1010], Partial Blast Results(1-1010)
Coverage 5x1010
[1-1010], 1-HCV-1a-partial - depth 5(1-1010)
"""
figure = build_coverage_figure(genome_coverage_csv)
assert expected_figure == summarize_figure(figure)
def test_plot_genome_coverage_partial():
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage
1-HCV-1a-partial,,1,,0,0,5
1-HCV-1a-partial,,2,,0,0,5
1-HCV-1a-partial,,3,,0,0,7
1-HCV-1a-partial,,4,,0,0,5
1-HCV-1a-partial,,5,,0,0,5
1-HCV-1a-partial,,6,,0,0,5
""")
expected_figure = """\
[1-500], Partial Blast Results(1-500)
Coverage 5x2, 7, 5x3
[1-6], 1-HCV-1a-partial - depth 7(1-500)
"""
figure = build_coverage_figure(genome_coverage_csv)
assert expected_figure == summarize_figure(figure)
def test_plot_genome_coverage_offset_blast():
""" When a contig extends before the reference start, offset everything. """
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,ins,dels,coverage
1-HCV-1a,HCV-1a,1,-2,0,0,5
1-HCV-1a,HCV-1a,2,-1,0,0,5
1-HCV-1a,HCV-1a,3,0,0,0,7
1-HCV-1a,HCV-1a,4,1,0,0,5
1-HCV-1a,HCV-1a,5,2,0,0,5
1-HCV-1a,HCV-1a,6,3,0,0,5
1-HCV-1a,HCV-1a,7,,0,0,5
1-HCV-1a,HCV-1a,8,,0,0,5
1-HCV-1a,HCV-1a,9,,0,0,5
1-HCV-1a,HCV-1a,10,4,0,0,5
1-HCV-1a,HCV-1a,11,5,0,0,5
1-HCV-1a,HCV-1a,12,6,0,0,5
2-unknown-partial,,1,,0,0,6
2-unknown-partial,,2,,0,0,6
2-unknown-partial,,3,,0,0,6
""")
blast_csv = StringIO("""\
contig_num,ref_name,score,match,pident,start,end,ref_start,ref_end
1,HCV-1a,30,0.33,90,10,12,4,6
""")
expected_figure = """\
5'[4-344], C[345-917], E1[918-1493], E2[1494-2582], p7[2583-2771], \
NS2[2772-3422], NS3[3423-5315], NS4b[5478-6260], NS4a[5316-5477], \
NS5a[6261-7604], NS5b[7605-9377], 3'[9378-9649]
7--1.1->9
7--1.1->9
Coverage 5x2, 7, 5x6
[1-9], 1-HCV-1a - depth 7(1-9649), lightgreen{7-9}
[4-503], [1004-1503], [2004-2503], [3004-3503], [4004-4503], \
[5004-5503], [6004-6503], [7004-7503], [8004-8503], [9004-9503], \
Partial Blast Results(4-9649)
Coverage 6x3
[4-6], 2-unknown-partial - depth 6(1-9649)
"""
figure = build_coverage_figure(genome_coverage_csv, blast_csv)
assert summarize_figure(figure) == expected_figure
def test_plot_genome_coverage_g2p():
genome_coverage_csv = StringIO("""\
contig,coordinates,query_nuc_pos,refseq_nuc_pos,dels,coverage,link
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,1,2261,0,5,M
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,2,2262,0,5,M
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,3,2263,0,5,M
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,4,2264,0,5,M
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,5,2265,0,5,M
1-HIV1-G-CM-KP718923-seed,HIV1-B-FR-K03455-seed,6,2266,0,5,M
HIV1-CON-XX-Consensus-seed,HIV1-B-FR-K03455-seed,1,7201,0,100,M
HIV1-CON-XX-Consensus-seed,HIV1-B-FR-K03455-seed,2,7202,0,100,M
HIV1-CON-XX-Consensus-seed,HIV1-B-FR-K03455-seed,3,7203,0,100,M
HIV1-CON-XX-Consensus-seed,HIV1-B-FR-K03455-seed,4,7204,0,100,M
HIV1-CON-XX-Consensus-seed,HIV1-B-FR-K03455-seed,5,7205,0,100,M
HIV1-CON-XX-Consensus-seed,HIV1-B-FR-K03455-seed,6,7206,0,100,M
""")
minimap_hits_csv = StringIO("""\
contig,ref_name,start,end,ref_start,ref_end
1-HIV1-G-CM-KP718923-seed,HIV1-G-CM-KP718923-seed,1,10,1653,1658
HIV1-CON-XX-Consensus-seed,HIV1-B-FR-K03455-seed,1,6,7201,7206
""")
expected_figure = """\
5' LTR[1-634], gag[789-2289], vif[5040-5616], tat[8379-8469], nef[8796-9414]
tat[5831-6045], vpu[6061-6307], rev[8379-8653], 3' LTR[9086-9719]
pol[2085-5096], vpr[5558-5847], rev[5970-6045], env[6225-8795]
PR[2252-2549], RT[2549-3869], INT[4229-5093], V3[7109-7217], GP41[7757-8792]
2261--1.1->2266
Coverage 100x6
[7201-7206], HIV1-CON-XX-Consensus-seed - depth 100(1-9719)
2261--1.1->2266
Coverage 5x6
[2261-2266], 1-HIV1-G-CM-KP718923-seed - depth 5(1-9719)
"""
figure = build_coverage_figure(genome_coverage_csv, minimap_hits_csv)
assert summarize_figure(figure) == expected_figure
def main():
args = parse_args()
source_bins = load_read_bins(args.remap1_csv)
target_bins = load_read_bins(args.remap2_csv)
all_keys = set(source_bins)
all_keys.update(target_bins)
moves = defaultdict(Counter) # {source_bin: {target_bin: count}}
for read_key in all_keys:
source_bin = source_bins.get(read_key)
target_bin = target_bins.get(read_key)
bin_moves = moves[source_bin]
bin_moves[target_bin] += 1
old_unmapped = moves.pop(None, None)
sorted_moves = sorted(moves.items())
if old_unmapped:
sorted_moves.insert(0, (None, old_unmapped))
diagram_path = args.compare_mapping_svg
f = build_coverage_figure(args.genome_coverage1_csv)
del f.elements[6:]
ref_y, ref = f.elements[5]
f2 = build_coverage_figure(args.genome_coverage2_csv)
coverage3_y, coverage3 = f2.elements[4]
contig3_y, contig3 = f2.elements[5]
coverage1_y, coverage1 = f2.elements[6]
contig1_y, contig1 = f2.elements[7]
dashes_y, dashes = f2.elements[10]
del f2.elements[4:]
# contig_y, contig = f2.elements[6]
# f.h += 50
f.w = max(f.w, f2.w)
f.add(coverage3, gap=-4)
f.add(contig3, gap=30)
contig3_y = f.elements[-1][0]
coverage1.a = 0
f.add(coverage1, gap=-4)
contig1_shift = contig1.tracks[0].a
for track in contig1.tracks:
if track.a >= contig1_shift:
track.a -= contig1_shift
track.b -= contig1_shift
f.add(contig1)
contig1_y = f.elements[-1][0]
if __name__ != '__live_coding__':
drawing_width = 970
else:
# noinspection PyProtectedMember
turtle_screen = Turtle._screen
drawing_width = turtle_screen.cv.cget('width') - 10
diagram_path = None
drawing = f.show(w=drawing_width)
seed1_y = f.h - f.elements[5][0] - 10
seed2_y = f.h - f.elements[7][0] + 25
ref_y = f.h - ref_y
contig_y = f.h - contig1_y + 25
contig_x = 0
x_scale = drawing_width / f.w
blast_display = BlastDisplay(drawing, x_scale, ref_y, contig_x, contig_y)
blast_rows = list(DictReader(args.blast2_csv))
blast_rows.sort(key=lambda match: int(match['score']), reverse=True)
best_ref = None
matched_positions = set()
for row in blast_rows:
if '003' not in row['contig_name']:
continue
if best_ref is None:
best_ref = row['ref_name']
elif row['ref_name'] != best_ref:
continue
start = int(row['start'])
end = int(row['end'])
new_positions = set(range(start, end))
collisions = matched_positions & new_positions
collision_fraction = len(collisions) / len(new_positions)
if collision_fraction < 0.1:
matched_positions |= new_positions
ref_start = int(row['ref_start'])
ref_end = int(row['ref_end'])
blast_display.add_match(start, end, ref_start, ref_end)
blast_display.draw()
for source_bin, bin_moves in sorted_moves:
total = sum(bin_moves.values())
for i, (target_bin, count) in enumerate(bin_moves.most_common()):
fraction = count / total
if fraction < 0.1:
break
if i == 0:
print(source_bin, end=':')
print(f'\t{target_bin}({fraction})')
source_shift = 0
target_shift = contig_x * x_scale
if source_bin is None:
source_x = target_bin
source_y = seed2_y - 10
source_shift = target_shift
else:
source_x = source_bin
source_y = seed1_y
if target_bin is None:
target_x = source_bin
target_y = seed1_y + 10
target_shift = source_shift
else:
target_x = target_bin
target_y = seed2_y
source_x *= 100 * x_scale
target_x *= 100 * x_scale
source_x += source_shift
target_x += target_shift
drawing.append(
Line(source_x,
source_y,
target_x,
target_y,
stroke='black',
stroke_width=2 * fraction,
stroke_opacity=0.25))
if diagram_path is not None:
drawing.saveSvg(diagram_path)
else:
display_image(drawing)