class DiagramTest(unittest.TestCase):
"""Creating feature sets, graph sets, tracks etc individually for the diagram."""
def setUp(self):
"""Test setup, just loads a GenBank file as a SeqRecord."""
handle = open(os.path.join("GenBank", "NC_005816.gb"), 'r')
self.record = SeqIO.read(handle, "genbank")
handle.close()
self.gdd = Diagram('Test Diagram')
# Add a track of features,
self.gdd.new_track(1, greytrack=True, name="CDS Features",
greytrack_labels=0, height=0.5)
def tearDown(self):
"""Release the drawing objects."""
del self.gdd
def test_str(self):
"""Test diagram's info as string."""
expected = "\n<<class 'Bio.Graphics.GenomeDiagram._Diagram.Diagram'>: Test Diagram>" \
"\n1 tracks" \
"\nTrack 1: " \
"\n<<class 'Bio.Graphics.GenomeDiagram._Track.Track'>: CDS Features>" \
"\n0 sets" \
"\n"
self.assertEqual(expected, str(self.gdd))
def test_add_track(self):
"""Add track."""
track = Track(name="Annotated Features")
self.gdd.add_track(track, 2)
self.assertEqual(2, len(self.gdd.get_tracks()))
def test_add_track_to_occupied_level(self):
"""Add track to occupied level."""
new_track = self.gdd.get_tracks()[0]
self.gdd.add_track(new_track, 1)
self.assertEqual(2, len(self.gdd.get_tracks()))
def test_add_track_error(self):
"""Test adding unspecified track."""
self.assertRaises(ValueError, self.gdd.add_track, None, 1)
def test_del_tracks(self):
"""Delete track."""
self.gdd.del_track(1)
self.assertEqual(0, len(self.gdd.get_tracks()))
def test_get_tracks(self):
"""Get track."""
self.assertEqual(1, len(self.gdd.get_tracks()))
def test_move_track(self):
"""Move a track."""
self.gdd.move_track(1, 2)
expected = "\n<<class 'Bio.Graphics.GenomeDiagram._Diagram.Diagram'>: Test Diagram>" \
"\n1 tracks" \
"\nTrack 2: " \
"\n<<class 'Bio.Graphics.GenomeDiagram._Track.Track'>: CDS Features>" \
"\n0 sets" \
"\n"
self.assertEqual(expected, str(self.gdd))
def test_renumber(self):
"""Test renumbering tracks."""
self.gdd.renumber_tracks(0)
expected = "\n<<class 'Bio.Graphics.GenomeDiagram._Diagram.Diagram'>: Test Diagram>" \
"\n1 tracks" \
"\nTrack 0: " \
"\n<<class 'Bio.Graphics.GenomeDiagram._Track.Track'>: CDS Features>" \
"\n0 sets" \
"\n"
self.assertEqual(expected, str(self.gdd))
def test_write_arguments(self):
"""Check how the write methods respond to output format arguments."""
gdd = Diagram('Test Diagram')
gdd.drawing = None # Hack - need the ReportLab drawing object to be created.
filename = os.path.join("Graphics", "error.txt")
# We (now) allow valid formats in any case.
for output in ["XXX", "xxx", None, 123, 5.9]:
with self.assertRaises(ValueError):
gdd.write(filename, output)
with self.assertRaises(ValueError):
gdd.write_to_string(output)
def test_partial_diagram(self):
"""Construct and draw SVG and PDF for just part of a SeqRecord."""
genbank_entry = self.record
start = 6500
end = 8750
gdd = Diagram('Test Diagram',
# For the circular diagram we don't want a closed cirle:
circular=False,
)
# Add a track of features,
gdt_features = gdd.new_track(1, greytrack=True,
name="CDS Features",
scale_largetick_interval=1000,
scale_smalltick_interval=100,
scale_format="SInt",
greytrack_labels=False,
height=0.5)
# We'll just use one feature set for these features,
gds_features = gdt_features.new_set()
for feature in genbank_entry.features:
if feature.type != "CDS":
# We're going to ignore these.
continue
if feature.location.end.position < start:
# Out of frame (too far left)
continue
if feature.location.start.position > end:
# Out of frame (too far right)
continue
# This URL should work in SVG output from recent versions
# of ReportLab. You need ReportLab 2.4 or later
try:
url = "http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi" +\
"?db=protein&id=%s" % feature.qualifiers["protein_id"][0]
except KeyError:
url = None
# Note that I am using strings for color names, instead
# of passing in color objects. This should also work!
if len(gds_features) % 2 == 0:
color = "white" # for testing the automatic black border!
else:
color = "red"
# Checking it can cope with the old UK spelling colour.
# Also show the labels perpendicular to the track.
gds_features.add_feature(feature, colour=color,
url=url,
sigil="ARROW",
label_position=None,
label_size=8,
label_angle=90,
label=True)
# And draw it...
gdd.draw(format='linear', orientation='landscape',
tracklines=False, pagesize=(10 * cm, 6 * cm), fragments=1,
start=start, end=end)
output_filename = os.path.join('Graphics', 'GD_region_linear.pdf')
gdd.write(output_filename, 'PDF')
# Also check the write_to_string (bytes string) method matches,
assert open(output_filename, "rb").read() == gdd.write_to_string('PDF')
output_filename = os.path.join('Graphics', 'GD_region_linear.svg')
gdd.write(output_filename, 'SVG')
# Circular with a particular start/end is a bit odd, but by setting
# circular=False (above) a sweep of 90% is used (a wedge is left out)
gdd.draw(format='circular',
tracklines=False, pagesize=(10 * cm, 10 * cm),
start=start, end=end)
output_filename = os.path.join('Graphics', 'GD_region_circular.pdf')
gdd.write(output_filename, 'PDF')
output_filename = os.path.join('Graphics', 'GD_region_circular.svg')
gdd.write(output_filename, 'SVG')
def test_diagram_via_methods_pdf(self):
"""Construct and draw PDF using method approach."""
genbank_entry = self.record
gdd = Diagram('Test Diagram')
# Add a track of features,
gdt_features = gdd.new_track(1, greytrack=True,
name="CDS Features", greytrack_labels=0,
height=0.5)
# We'll just use one feature set for the genes and misc_features,
gds_features = gdt_features.new_set()
for feature in genbank_entry.features:
if feature.type == "gene":
if len(gds_features) % 2 == 0:
color = "blue"
else:
color = "lightblue"
gds_features.add_feature(feature, color=color,
# label_position="middle",
# label_position="end",
label_position="start",
label_size=11,
# label_angle=90,
sigil="ARROW",
label=True)
# I want to include some strandless features, so for an example
# will use EcoRI recognition sites etc.
for site, name, color in [("GAATTC", "EcoRI", "green"),
("CCCGGG", "SmaI", "orange"),
("AAGCTT", "HindIII", "red"),
("GGATCC", "BamHI", "purple")]:
index = 0
while True:
index = genbank_entry.seq.find(site, start=index)
if index == -1:
break
feature = SeqFeature(FeatureLocation(index, index + 6), strand=None)
# This URL should work in SVG output from recent versions
# of ReportLab. You need ReportLab 2.4 or later
try:
url = "http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi" +\
"?db=protein&id=%s" % feature.qualifiers["protein_id"][0]
except KeyError:
url = None
gds_features.add_feature(feature, color=color,
url=url,
# label_position="middle",
label_size=10,
label_color=color,
# label_angle=90,
name=name,
label=True)
index += len(site)
del index
# Now add a graph track...
gdt_at_gc = gdd.new_track(2, greytrack=True,
name="AT and GC content",
greytrack_labels=True)
gds_at_gc = gdt_at_gc.new_set(type="graph")
step = len(genbank_entry) // 200
gds_at_gc.new_graph(apply_to_window(genbank_entry.seq, step, calc_gc_content, step),
'GC content', style='line',
color=colors.lightgreen,
altcolor=colors.darkseagreen)
gds_at_gc.new_graph(apply_to_window(genbank_entry.seq, step, calc_at_content, step),
'AT content', style='line',
color=colors.orange,
altcolor=colors.red)
# Finally draw it in both formats,
gdd.draw(format='linear', orientation='landscape', tracklines=0,
pagesize='A4', fragments=3)
output_filename = os.path.join('Graphics', 'GD_by_meth_linear.pdf')
gdd.write(output_filename, 'PDF')
gdd.draw(format='circular', tracklines=False, circle_core=0.8,
pagesize=(20 * cm, 20 * cm), circular=True)
output_filename = os.path.join('Graphics', 'GD_by_meth_circular.pdf')
gdd.write(output_filename, 'PDF')
def test_diagram_via_object_pdf(self):
"""Construct and draw PDF using object approach."""
genbank_entry = self.record
gdd = Diagram('Test Diagram')
gdt1 = Track('CDS features', greytrack=True,
scale_largetick_interval=1e4,
scale_smalltick_interval=1e3,
greytrack_labels=10,
greytrack_font_color="red",
scale_format="SInt")
gdt2 = Track('gene features', greytrack=1, scale_largetick_interval=1e4)
# First add some feature sets:
gdfsA = FeatureSet(name='CDS backgrounds')
gdfsB = FeatureSet(name='gene background')
gdfs1 = FeatureSet(name='CDS features')
gdfs2 = FeatureSet(name='gene features')
gdfs3 = FeatureSet(name='misc_features')
gdfs4 = FeatureSet(name='repeat regions')
prev_gene = None
cds_count = 0
for feature in genbank_entry.features:
if feature.type == 'CDS':
cds_count += 1
if prev_gene:
# Assuming it goes with this CDS!
if cds_count % 2 == 0:
dark, light = colors.peru, colors.tan
else:
dark, light = colors.burlywood, colors.bisque
# Background for CDS,
a = gdfsA.add_feature(SeqFeature(FeatureLocation(feature.location.start, feature.location.end, strand=0)),
color=dark)
# Background for gene,
b = gdfsB.add_feature(SeqFeature(FeatureLocation(prev_gene.location.start, prev_gene.location.end, strand=0)),
color=dark)
# Cross link,
gdd.cross_track_links.append(CrossLink(a, b, light, dark))
prev_gene = None
if feature.type == 'gene':
prev_gene = feature
# Some cross links on the same linear diagram fragment,
f, c = fill_and_border(colors.red)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(2220, 2230)), color=f, border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(2200, 2210)), color=f, border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c))
f, c = fill_and_border(colors.blue)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(2150, 2200)), color=f, border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(2220, 2290)), color=f, border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c, flip=True))
f, c = fill_and_border(colors.green)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(2250, 2560)), color=f, border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(2300, 2860)), color=f, border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c))
# Some cross links where both parts are saddling the linear diagram fragment boundary,
f, c = fill_and_border(colors.red)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(3155, 3250)), color=f, border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(3130, 3300)), color=f, border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c))
# Nestled within that (drawn on top),
f, c = fill_and_border(colors.blue)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(3160, 3275)), color=f, border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(3180, 3225)), color=f, border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c, flip=True))
# Some cross links where two features are on either side of the linear diagram fragment boundary,
f, c = fill_and_border(colors.green)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(6450, 6550)), color=f, border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(6265, 6365)), color=f, border=c)
gdd.cross_track_links.append(CrossLink(a, b, color=f, border=c))
f, c = fill_and_border(colors.gold)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(6265, 6365)), color=f, border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(6450, 6550)), color=f, border=c)
gdd.cross_track_links.append(CrossLink(a, b, color=f, border=c))
f, c = fill_and_border(colors.red)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(6275, 6375)), color=f, border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(6430, 6530)), color=f, border=c)
gdd.cross_track_links.append(CrossLink(a, b, color=f, border=c, flip=True))
f, c = fill_and_border(colors.blue)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(6430, 6530)), color=f, border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(6275, 6375)), color=f, border=c)
gdd.cross_track_links.append(CrossLink(a, b, color=f, border=c, flip=True))
cds_count = 0
for feature in genbank_entry.features:
if feature.type == 'CDS':
cds_count += 1
if cds_count % 2 == 0:
gdfs1.add_feature(feature, color=colors.pink, sigil="ARROW")
else:
gdfs1.add_feature(feature, color=colors.red, sigil="ARROW")
if feature.type == 'gene':
# Note we set the colour of ALL the genes later on as a test,
gdfs2.add_feature(feature, sigil="ARROW")
if feature.type == 'misc_feature':
gdfs3.add_feature(feature, color=colors.orange)
if feature.type == 'repeat_region':
gdfs4.add_feature(feature, color=colors.purple)
# gdd.cross_track_links = gdd.cross_track_links[:1]
gdfs1.set_all_features('label', 1)
gdfs2.set_all_features('label', 1)
gdfs3.set_all_features('label', 1)
gdfs4.set_all_features('label', 1)
gdfs3.set_all_features('hide', 0)
gdfs4.set_all_features('hide', 0)
# gdfs1.set_all_features('color', colors.red)
gdfs2.set_all_features('color', colors.blue)
gdt1.add_set(gdfsA) # Before CDS so under them!
gdt1.add_set(gdfs1)
gdt2.add_set(gdfsB) # Before genes so under them!
gdt2.add_set(gdfs2)
gdt3 = Track('misc features and repeats', greytrack=1,
scale_largetick_interval=1e4)
gdt3.add_set(gdfs3)
gdt3.add_set(gdfs4)
# Now add some graph sets:
# Use a fairly large step so we can easily tell the difference
# between the bar and line graphs.
step = len(genbank_entry) // 200
gdgs1 = GraphSet('GC skew')
graphdata1 = apply_to_window(genbank_entry.seq, step, calc_gc_skew, step)
gdgs1.new_graph(graphdata1, 'GC Skew', style='bar',
color=colors.violet, altcolor=colors.purple)
gdt4 = Track('GC Skew (bar)', height=1.94, greytrack=1,
scale_largetick_interval=1e4)
gdt4.add_set(gdgs1)
gdgs2 = GraphSet('GC and AT Content')
gdgs2.new_graph(apply_to_window(genbank_entry.seq, step, calc_gc_content, step),
'GC content', style='line', color=colors.lightgreen,
altcolor=colors.darkseagreen)
gdgs2.new_graph(apply_to_window(genbank_entry.seq, step, calc_at_content, step),
'AT content', style='line', color=colors.orange,
altcolor=colors.red)
gdt5 = Track('GC Content(green line), AT Content(red line)',
height=1.94, greytrack=1, scale_largetick_interval=1e4)
gdt5.add_set(gdgs2)
gdgs3 = GraphSet('Di-nucleotide count')
step = len(genbank_entry) // 400 # smaller step
gdgs3.new_graph(apply_to_window(genbank_entry.seq, step, calc_dinucleotide_counts, step),
'Di-nucleotide count', style='heat',
color=colors.red, altcolor=colors.orange)
gdt6 = Track('Di-nucleotide count', height=0.5, greytrack=False, scale=False)
gdt6.add_set(gdgs3)
# Add the tracks (from both features and graphs)
# Leave some white space in the middle/bottom
gdd.add_track(gdt4, 3) # GC skew
gdd.add_track(gdt5, 4) # GC and AT content
gdd.add_track(gdt1, 5) # CDS features
gdd.add_track(gdt2, 6) # Gene features
gdd.add_track(gdt3, 7) # Misc features and repeat feature
gdd.add_track(gdt6, 8) # Feature depth
# Finally draw it in both formats, and full view and partial
gdd.draw(format='circular', orientation='landscape',
tracklines=0, pagesize='A0')
output_filename = os.path.join('Graphics', 'GD_by_obj_circular.pdf')
gdd.write(output_filename, 'PDF')
gdd.circular = False
gdd.draw(format='circular', orientation='landscape',
tracklines=0, pagesize='A0', start=3000, end=6300)
output_filename = os.path.join('Graphics', 'GD_by_obj_frag_circular.pdf')
gdd.write(output_filename, 'PDF')
gdd.draw(format='linear', orientation='landscape',
tracklines=0, pagesize='A0', fragments=3)
output_filename = os.path.join('Graphics', 'GD_by_obj_linear.pdf')
gdd.write(output_filename, 'PDF')
gdd.set_all_tracks("greytrack_labels", 2)
gdd.draw(format='linear', orientation='landscape',
tracklines=0, pagesize=(30 * cm, 10 * cm), fragments=1,
start=3000, end=6300)
output_filename = os.path.join('Graphics', 'GD_by_obj_frag_linear.pdf')
gdd.write(output_filename, 'PDF')
class DiagramTest(unittest.TestCase):
"""Creating feature sets, graph sets, tracks etc individually for the diagram."""
def setUp(self):
"""Test setup, just loads a GenBank file as a SeqRecord."""
handle = open(os.path.join("GenBank", "NC_005816.gb"), 'r')
self.record = SeqIO.read(handle, "genbank")
handle.close()
self.gdd = Diagram('Test Diagram')
# Add a track of features,
self.gdd.new_track(1,
greytrack=True,
name="CDS Features",
greytrack_labels=0,
height=0.5)
def tearDown(self):
del self.gdd
def test_str(self):
"""Test diagram's info as string."""
expected = "\n<<class 'Bio.Graphics.GenomeDiagram._Diagram.Diagram'>: Test Diagram>" \
"\n1 tracks" \
"\nTrack 1: " \
"\n<<class 'Bio.Graphics.GenomeDiagram._Track.Track'>: CDS Features>" \
"\n0 sets" \
"\n"
self.assertEqual(expected, str(self.gdd))
def test_add_track(self):
track = Track(name="Annotated Features")
self.gdd.add_track(track, 2)
self.assertEqual(2, len(self.gdd.get_tracks()))
def test_add_track_to_occupied_level(self):
new_track = self.gdd.get_tracks()[0]
self.gdd.add_track(new_track, 1)
self.assertEqual(2, len(self.gdd.get_tracks()))
def test_add_track_error(self):
"""Test adding unspecified track."""
self.assertRaises(ValueError, self.gdd.add_track, None, 1)
def test_del_tracks(self):
self.gdd.del_track(1)
self.assertEqual(0, len(self.gdd.get_tracks()))
def test_get_tracks(self):
self.assertEqual(1, len(self.gdd.get_tracks()))
def test_move_track(self):
self.gdd.move_track(1, 2)
expected = "\n<<class 'Bio.Graphics.GenomeDiagram._Diagram.Diagram'>: Test Diagram>" \
"\n1 tracks" \
"\nTrack 2: " \
"\n<<class 'Bio.Graphics.GenomeDiagram._Track.Track'>: CDS Features>" \
"\n0 sets" \
"\n"
self.assertEqual(expected, str(self.gdd))
def test_renumber(self):
"""Test renumbering tracks."""
self.gdd.renumber_tracks(0)
expected = "\n<<class 'Bio.Graphics.GenomeDiagram._Diagram.Diagram'>: Test Diagram>" \
"\n1 tracks" \
"\nTrack 0: " \
"\n<<class 'Bio.Graphics.GenomeDiagram._Track.Track'>: CDS Features>" \
"\n0 sets" \
"\n"
self.assertEqual(expected, str(self.gdd))
def test_write_arguments(self):
"""Check how the write methods respond to output format arguments."""
gdd = Diagram('Test Diagram')
gdd.drawing = None # Hack - need the ReportLab drawing object to be created.
filename = os.path.join("Graphics", "error.txt")
# We (now) allow valid formats in any case.
for output in ["XXX", "xxx", None, 123, 5.9]:
try:
gdd.write(filename, output)
assert False, \
"Should have rejected %s as an output format" % output
except ValueError:
# Good!
pass
try:
gdd.write_to_string(output)
assert False, \
"Should have rejected %s as an output format" % output
except ValueError:
# Good!
pass
def test_partial_diagram(self):
"""construct and draw SVG and PDF for just part of a SeqRecord."""
genbank_entry = self.record
start = 6500
end = 8750
gdd = Diagram(
'Test Diagram',
# For the circular diagram we don't want a closed cirle:
circular=False,
)
# Add a track of features,
gdt_features = gdd.new_track(1,
greytrack=True,
name="CDS Features",
scale_largetick_interval=1000,
scale_smalltick_interval=100,
scale_format="SInt",
greytrack_labels=False,
height=0.5)
# We'll just use one feature set for these features,
gds_features = gdt_features.new_set()
for feature in genbank_entry.features:
if feature.type != "CDS":
# We're going to ignore these.
continue
if feature.location.end.position < start:
# Out of frame (too far left)
continue
if feature.location.start.position > end:
# Out of frame (too far right)
continue
# This URL should work in SVG output from recent versions
# of ReportLab. You need ReportLab 2.4 or later
try:
url = "http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi" +\
"?db=protein&id=%s" % feature.qualifiers["protein_id"][0]
except KeyError:
url = None
# Note that I am using strings for color names, instead
# of passing in color objects. This should also work!
if len(gds_features) % 2 == 0:
color = "white" # for testing the automatic black border!
else:
color = "red"
# Checking it can cope with the old UK spelling colour.
# Also show the labels perpendicular to the track.
gds_features.add_feature(feature,
colour=color,
url=url,
sigil="ARROW",
label_position=None,
label_size=8,
label_angle=90,
label=True)
# And draw it...
gdd.draw(format='linear',
orientation='landscape',
tracklines=False,
pagesize=(10 * cm, 6 * cm),
fragments=1,
start=start,
end=end)
output_filename = os.path.join('Graphics', 'GD_region_linear.pdf')
gdd.write(output_filename, 'PDF')
# Also check the write_to_string (bytes string) method matches,
# (Note the possible confusion over new lines on Windows)
assert open(output_filename, "rb").read().replace(b"\r\n", b"\n") \
== gdd.write_to_string('PDF').replace(b"\r\n", b"\n")
output_filename = os.path.join('Graphics', 'GD_region_linear.svg')
gdd.write(output_filename, 'SVG')
# Circular with a particular start/end is a bit odd, but by setting
# circular=False (above) a sweep of 90% is used (a wedge is left out)
gdd.draw(format='circular',
tracklines=False,
pagesize=(10 * cm, 10 * cm),
start=start,
end=end)
output_filename = os.path.join('Graphics', 'GD_region_circular.pdf')
gdd.write(output_filename, 'PDF')
output_filename = os.path.join('Graphics', 'GD_region_circular.svg')
gdd.write(output_filename, 'SVG')
def test_diagram_via_methods_pdf(self):
"""Construct and draw PDF using method approach."""
genbank_entry = self.record
gdd = Diagram('Test Diagram')
# Add a track of features,
gdt_features = gdd.new_track(1,
greytrack=True,
name="CDS Features",
greytrack_labels=0,
height=0.5)
# We'll just use one feature set for the genes and misc_features,
gds_features = gdt_features.new_set()
for feature in genbank_entry.features:
if feature.type == "gene":
if len(gds_features) % 2 == 0:
color = "blue"
else:
color = "lightblue"
gds_features.add_feature(
feature,
color=color,
# label_position="middle",
# label_position="end",
label_position="start",
label_size=11,
# label_angle=90,
sigil="ARROW",
label=True)
# I want to include some strandless features, so for an example
# will use EcoRI recognition sites etc.
for site, name, color in [("GAATTC", "EcoRI", "green"),
("CCCGGG", "SmaI", "orange"),
("AAGCTT", "HindIII", "red"),
("GGATCC", "BamHI", "purple")]:
index = 0
while True:
index = genbank_entry.seq.find(site, start=index)
if index == -1:
break
feature = SeqFeature(FeatureLocation(index, index + 6),
strand=None)
# This URL should work in SVG output from recent versions
# of ReportLab. You need ReportLab 2.4 or later
try:
url = "http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi" +\
"?db=protein&id=%s" % feature.qualifiers["protein_id"][0]
except KeyError:
url = None
gds_features.add_feature(
feature,
color=color,
url=url,
# label_position="middle",
label_size=10,
label_color=color,
# label_angle=90,
name=name,
label=True)
index += len(site)
del index
# Now add a graph track...
gdt_at_gc = gdd.new_track(2,
greytrack=True,
name="AT and GC content",
greytrack_labels=True)
gds_at_gc = gdt_at_gc.new_set(type="graph")
step = len(genbank_entry) // 200
gds_at_gc.new_graph(apply_to_window(genbank_entry.seq, step,
calc_gc_content, step),
'GC content',
style='line',
color=colors.lightgreen,
altcolor=colors.darkseagreen)
gds_at_gc.new_graph(apply_to_window(genbank_entry.seq, step,
calc_at_content, step),
'AT content',
style='line',
color=colors.orange,
altcolor=colors.red)
# Finally draw it in both formats,
gdd.draw(format='linear',
orientation='landscape',
tracklines=0,
pagesize='A4',
fragments=3)
output_filename = os.path.join('Graphics', 'GD_by_meth_linear.pdf')
gdd.write(output_filename, 'PDF')
gdd.draw(format='circular',
tracklines=False,
circle_core=0.8,
pagesize=(20 * cm, 20 * cm),
circular=True)
output_filename = os.path.join('Graphics', 'GD_by_meth_circular.pdf')
gdd.write(output_filename, 'PDF')
def test_diagram_via_object_pdf(self):
"""Construct and draw PDF using object approach."""
genbank_entry = self.record
gdd = Diagram('Test Diagram')
gdt1 = Track('CDS features',
greytrack=True,
scale_largetick_interval=1e4,
scale_smalltick_interval=1e3,
greytrack_labels=10,
greytrack_font_color="red",
scale_format="SInt")
gdt2 = Track('gene features',
greytrack=1,
scale_largetick_interval=1e4)
# First add some feature sets:
gdfsA = FeatureSet(name='CDS backgrounds')
gdfsB = FeatureSet(name='gene background')
gdfs1 = FeatureSet(name='CDS features')
gdfs2 = FeatureSet(name='gene features')
gdfs3 = FeatureSet(name='misc_features')
gdfs4 = FeatureSet(name='repeat regions')
prev_gene = None
cds_count = 0
for feature in genbank_entry.features:
if feature.type == 'CDS':
cds_count += 1
if prev_gene:
# Assuming it goes with this CDS!
if cds_count % 2 == 0:
dark, light = colors.peru, colors.tan
else:
dark, light = colors.burlywood, colors.bisque
# Background for CDS,
a = gdfsA.add_feature(SeqFeature(
FeatureLocation(feature.location.start,
feature.location.end,
strand=0)),
color=dark)
# Background for gene,
b = gdfsB.add_feature(SeqFeature(
FeatureLocation(prev_gene.location.start,
prev_gene.location.end,
strand=0)),
color=dark)
# Cross link,
gdd.cross_track_links.append(CrossLink(a, b, light, dark))
prev_gene = None
if feature.type == 'gene':
prev_gene = feature
# Some cross links on the same linear diagram fragment,
f, c = fill_and_border(colors.red)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(2220, 2230)),
color=f,
border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(2200, 2210)),
color=f,
border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c))
f, c = fill_and_border(colors.blue)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(2150, 2200)),
color=f,
border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(2220, 2290)),
color=f,
border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c, flip=True))
f, c = fill_and_border(colors.green)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(2250, 2560)),
color=f,
border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(2300, 2860)),
color=f,
border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c))
# Some cross links where both parts are saddling the linear diagram fragment boundary,
f, c = fill_and_border(colors.red)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(3155, 3250)),
color=f,
border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(3130, 3300)),
color=f,
border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c))
# Nestled within that (drawn on top),
f, c = fill_and_border(colors.blue)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(3160, 3275)),
color=f,
border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(3180, 3225)),
color=f,
border=c)
gdd.cross_track_links.append(CrossLink(a, b, f, c, flip=True))
# Some cross links where two features are on either side of the linear diagram fragment boundary,
f, c = fill_and_border(colors.green)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(6450, 6550)),
color=f,
border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(6265, 6365)),
color=f,
border=c)
gdd.cross_track_links.append(CrossLink(a, b, color=f, border=c))
f, c = fill_and_border(colors.gold)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(6265, 6365)),
color=f,
border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(6450, 6550)),
color=f,
border=c)
gdd.cross_track_links.append(CrossLink(a, b, color=f, border=c))
f, c = fill_and_border(colors.red)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(6275, 6375)),
color=f,
border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(6430, 6530)),
color=f,
border=c)
gdd.cross_track_links.append(
CrossLink(a, b, color=f, border=c, flip=True))
f, c = fill_and_border(colors.blue)
a = gdfsA.add_feature(SeqFeature(FeatureLocation(6430, 6530)),
color=f,
border=c)
b = gdfsB.add_feature(SeqFeature(FeatureLocation(6275, 6375)),
color=f,
border=c)
gdd.cross_track_links.append(
CrossLink(a, b, color=f, border=c, flip=True))
cds_count = 0
for feature in genbank_entry.features:
if feature.type == 'CDS':
cds_count += 1
if cds_count % 2 == 0:
gdfs1.add_feature(feature,
color=colors.pink,
sigil="ARROW")
else:
gdfs1.add_feature(feature, color=colors.red, sigil="ARROW")
if feature.type == 'gene':
# Note we set the colour of ALL the genes later on as a test,
gdfs2.add_feature(feature, sigil="ARROW")
if feature.type == 'misc_feature':
gdfs3.add_feature(feature, color=colors.orange)
if feature.type == 'repeat_region':
gdfs4.add_feature(feature, color=colors.purple)
# gdd.cross_track_links = gdd.cross_track_links[:1]
gdfs1.set_all_features('label', 1)
gdfs2.set_all_features('label', 1)
gdfs3.set_all_features('label', 1)
gdfs4.set_all_features('label', 1)
gdfs3.set_all_features('hide', 0)
gdfs4.set_all_features('hide', 0)
# gdfs1.set_all_features('color', colors.red)
gdfs2.set_all_features('color', colors.blue)
gdt1.add_set(gdfsA) # Before CDS so under them!
gdt1.add_set(gdfs1)
gdt2.add_set(gdfsB) # Before genes so under them!
gdt2.add_set(gdfs2)
gdt3 = Track('misc features and repeats',
greytrack=1,
scale_largetick_interval=1e4)
gdt3.add_set(gdfs3)
gdt3.add_set(gdfs4)
# Now add some graph sets:
# Use a fairly large step so we can easily tell the difference
# between the bar and line graphs.
step = len(genbank_entry) // 200
gdgs1 = GraphSet('GC skew')
graphdata1 = apply_to_window(genbank_entry.seq, step, calc_gc_skew,
step)
gdgs1.new_graph(graphdata1,
'GC Skew',
style='bar',
color=colors.violet,
altcolor=colors.purple)
gdt4 = Track('GC Skew (bar)',
height=1.94,
greytrack=1,
scale_largetick_interval=1e4)
gdt4.add_set(gdgs1)
gdgs2 = GraphSet('GC and AT Content')
gdgs2.new_graph(apply_to_window(genbank_entry.seq, step,
calc_gc_content, step),
'GC content',
style='line',
color=colors.lightgreen,
altcolor=colors.darkseagreen)
gdgs2.new_graph(apply_to_window(genbank_entry.seq, step,
calc_at_content, step),
'AT content',
style='line',
color=colors.orange,
altcolor=colors.red)
gdt5 = Track('GC Content(green line), AT Content(red line)',
height=1.94,
greytrack=1,
scale_largetick_interval=1e4)
gdt5.add_set(gdgs2)
gdgs3 = GraphSet('Di-nucleotide count')
step = len(genbank_entry) // 400 # smaller step
gdgs3.new_graph(apply_to_window(genbank_entry.seq, step,
calc_dinucleotide_counts, step),
'Di-nucleotide count',
style='heat',
color=colors.red,
altcolor=colors.orange)
gdt6 = Track('Di-nucleotide count',
height=0.5,
greytrack=False,
scale=False)
gdt6.add_set(gdgs3)
# Add the tracks (from both features and graphs)
# Leave some white space in the middle/bottom
gdd.add_track(gdt4, 3) # GC skew
gdd.add_track(gdt5, 4) # GC and AT content
gdd.add_track(gdt1, 5) # CDS features
gdd.add_track(gdt2, 6) # Gene features
gdd.add_track(gdt3, 7) # Misc features and repeat feature
gdd.add_track(gdt6, 8) # Feature depth
# Finally draw it in both formats, and full view and partial
gdd.draw(format='circular',
orientation='landscape',
tracklines=0,
pagesize='A0')
output_filename = os.path.join('Graphics', 'GD_by_obj_circular.pdf')
gdd.write(output_filename, 'PDF')
gdd.circular = False
gdd.draw(format='circular',
orientation='landscape',
tracklines=0,
pagesize='A0',
start=3000,
end=6300)
output_filename = os.path.join('Graphics',
'GD_by_obj_frag_circular.pdf')
gdd.write(output_filename, 'PDF')
gdd.draw(format='linear',
orientation='landscape',
tracklines=0,
pagesize='A0',
fragments=3)
output_filename = os.path.join('Graphics', 'GD_by_obj_linear.pdf')
gdd.write(output_filename, 'PDF')
gdd.set_all_tracks("greytrack_labels", 2)
gdd.draw(format='linear',
orientation='landscape',
tracklines=0,
pagesize=(30 * cm, 10 * cm),
fragments=1,
start=3000,
end=6300)
output_filename = os.path.join('Graphics', 'GD_by_obj_frag_linear.pdf')
gdd.write(output_filename, 'PDF')
