def _compose_crosslink(self, index, result, features1, features2):
for f1, f2, col in result:
if f1 is None: continue
if self.no_color: col = colors.color2bw(col)
tf1 = features1.add_feature(f1, color=col, border=col)
tf2 = features2.add_feature(f2, color=col, border=col)
self.diagram.cross_track_links.append(CrossLink(tf1, tf2, col, col))
flip = not flip
q_start, q_end = q_end, q_start
if s_start > s_end:
flip = not flip
s_start, s_end = s_end, s_start
if flip:
c = colors.Color(0, 0, 1, alpha=0.25)
b = False
else:
c = colors.Color(1, 0, 0, alpha=0.25)
b = False
q_feature = q_set.add_feature(SeqFeature(FeatureLocation(q_start-1, q_end)),
color=c, border=b)
s_feature = s_set.add_feature(SeqFeature(FeatureLocation(s_start-1, s_end)),
color=c, border=b)
gd_diagram.cross_track_links.append(CrossLink(q_feature, s_feature, c, b))
#NOTE: We are using the same colour for all the matches,
#with transparency. This means overlayed matches will appear darker.
#It also means the drawing order not very important.
#Note ACT puts long hits at the back, and colours by hit score
handle.close()
print("Drawing CDS features...")
for f, format in genomes:
record = records[f]
feature_set = feature_sets[f]
#Mark the CDS features
for cds in record.features:
if cds.type != "CDS":
continue
feature_set.add_feature(cds, sigil="ARROW",
if flip:
c = colors.Color(0, 0, 1, alpha=0.25)
b = False
else:
c = colors.Color(1, 0, 0, alpha=0.25)
b = False
q_feature = q_set.add_feature(SeqFeature(
FeatureLocation(q_start - 1, q_end)),
color=c,
border=b)
s_feature = s_set.add_feature(SeqFeature(
FeatureLocation(s_start - 1, s_end)),
color=c,
border=b)
gd_diagram.cross_track_links.append(
CrossLink(q_feature, s_feature, c, b))
#NOTE: We are using the same colour for all the matches,
#with transparency. This means overlayed matches will appear darker.
#It also means the drawing order not very important.
#Note ACT puts long hits at the back, and colours by hit score
handle.close()
print("Drawing CDS features...")
for f, format in genomes:
record = records[f]
feature_set = feature_sets[f]
#Mark the CDS features
for cds in record.features:
if cds.type != "CDS":
continue
feature_set.add_feature(cds,
def crosslinks(fileName, GenBank_1, GenBank_2):
gd_diagram = GenomeDiagram.Diagram(fileName)
max_len = 0
#Open Files and create fasta files to be compared by Ublast
A_rec = SeqIO.read(GenBank_1, 'gb')
GB_file_name = GenBank_1
fasta_file_name_A = GB_file_name[:-3] + ".fasta"
writeFasta(GB_file_name, fasta_file_name_A)
B_rec = SeqIO.read(GenBank_2, 'gb')
GB_file_name = GenBank_2
fasta_file_name_B = GB_file_name[:-3] + ".fasta"
writeFasta(GB_file_name, fasta_file_name_B)
#create the tab file with the Ublast output
blastfile = ublastfeatures(fasta_file_name_A, fasta_file_name_B)
Gname = 'nn' #name of gene to add
#First section gets the crosslinks from the blast files
A_vs_B = getCrossLinks(blastfile)
#print ('(percent, Gene Query, Gene result)')#This prints the list of Blast results for reference
for item in A_vs_B:
print item
# asks user for a gene name to highlight
gene_search = raw_input(
"would you like to highlight a specific gene name?\n \t1) Yes\n \t2) No\n"
)
if gene_search == "1" or gene_search.lower() == "yes":
gene_highlight = raw_input(
"What is the name of the gene you would like to highlight?\n")
print gene_highlight + " will be printed in red on the genome diagram, all other genes will be grey"
C_colors = [yellow] * 1 + [orange] * 1 + [brown] * 1 + [
lightblue
] * 1 + [purple] * 1 + [green] * 1 + [grey] * 1
else:
gene_highlight = "NONE"
C_colors = [yellow] * 1 + [
orange
] * 1 + [brown] * 1 + [lightblue] * 1 + [purple] * 1 + [green] * 1 + [
grey
] * 1 #this creates an array of color for the arrows in the GUI
i = 0 #index of random color to add
geneColor = grey #color of gene. Grey= no name
# Create new features for concatenations
recs = ("A", "B")
for rec in recs:
if rec == "A":
for loc_a in re.finditer('NNNNNCACACACTTAATTAATTAAGTGTGTGNNNNN',
str(A_rec.seq)):
concat_feature = SeqFeature(FeatureLocation(loc_a.start(),
loc_a.start() + 35,
strand=-1),
id="Concat",
type="CDS",
qualifiers={'product': 'Concat'})
A_rec.features.append(concat_feature)
else:
for loc_b in re.finditer('NNNNNCACACACTTAATTAATTAAGTGTGTGNNNNN',
str(B_rec.seq)):
concat_feature = SeqFeature(FeatureLocation(loc_b.start(),
loc_b.start() + 35,
strand=-1),
id="Concat",
type="CDS",
qualifiers={'product': 'Concat'})
B_rec.features.append(concat_feature)
#Read in lists of gene names and types
with open('Backbones_2_Clean.csv', 'r') as hand1:
back_b = csv.reader(hand1)
backbone = list(back_b)
with open('AntibioticResistanceGenesClean.csv', 'r') as hand2:
AnRe = csv.reader(hand2)
An_Re = list(AnRe)
#this loop adds each gene feature to the record with a color and name
for record, gene_colors in zip([A_rec, B_rec], [C_colors, C_colors]):
max_len = max(max_len, len(record))
gd_track_for_features = gd_diagram.new_track(1,
name=record.name,
greytrack=True,
start=0,
end=len(record))
gd_feature_set = gd_track_for_features.new_set()
for feature in record.features:
if feature.type != "CDS":
#Exclude this feature
continue
## Chose Colors of annotations based on gene name
try:
Gname = feature.qualifiers['product'][0]
if Gname == gene_highlight:
geneColor = red
elif gene_highlight == "NONE":
geneColor = gene_colors[i % 6]
# Backbone genes
elif Gname in backbone[0]:
geneColor = blue
#Antibiotic Resistance Genes
elif Gname in An_Re[0]:
geneColor = green
#Transposease & Intergrase
elif Gname == 'Tnp' or Gname == 'Int':
geneColor = orange
else:
geneColor = grey
except KeyError: #if no gene name make it grey
Gname = 'No Name'
geneColor = grey
gd_feature_set.add_feature(
feature,
sigil="BIGARROW", #this adds gene features to gd_feature_set
arrowhead_length=.25,
color=geneColor,
label=True,
name=Gname,
label_position="start",
label_size=6,
label_angle=45)
i += 1 #increment i so that arrows will have a random color
track_X = gd_diagram.tracks[2]
track_Y = gd_diagram.tracks[1]
#this loop adds the cross links so they point to their feature in the diagram
for score, id_X, id_Y in A_vs_B:
try:
feature_X = get_feature(A_rec.features, id_X)
feature_Y = get_feature(B_rec.features, id_Y)
color = colors.linearlyInterpolatedColor(colors.white,
colors.firebrick, 0, 100,
score)
link_xy = CrossLink(
(track_X, feature_X.location.start, feature_X.location.end),
(track_Y, feature_Y.location.start, feature_Y.location.end),
color, colors.lightgrey)
print "Link made "
gd_diagram.cross_track_links.append(link_xy)
except KeyError:
print "Feature qualifier for crosslink not found" # for those pesky nameless genes
gd_diagram.draw(format="linear",
pagesize=(1200, 2400),
fragments=1,
start=0,
end=max_len)
print max_len
gd_diagram.write(fileName + ".pdf", "PDF")
break
# ==========================================================================
# Build links
# ==========================================================================
for id, link in links.iteritems():
if id in merged_links:
continue
feat1 = features[link["fid1"]]
feat2 = features[link["fid2"]]
'''
if "NapDC1_B04" not in name_mapping[feat1["cid"]] and "NapDC1_B04" not in name_mapping[feat2["cid"]]:
continue
'''
color = colors.Color(50, 50, 50, alpha=0)
gd_link = CrossLink((link["track1"], link["start1"], link["end1"]),
(link["track2"], link["start2"], link["end2"]),
color, colors.lightgrey, feat1["dir"] != feat2["dir"])
gd_diagram.cross_track_links.append(gd_link)
print "Number of links: " + str(len(gd_diagram.cross_track_links))
# ==========================================================================
# Write to file
# ==========================================================================
gd_diagram.draw(format="linear", pagesize="A1")
gd_diagram.write(mapping_filename + ".svg", output="SVG", dpi=600)
alpha=(hsp.ident_pct / 200.0))
loc = FeatureLocation(offset + hsp.query_start,
offset + hsp.query_end,
strand=0)
q = gd_contig_features.add_feature(SeqFeature(loc),
color=color,
border=border)
r_offset = ref_offsets[hack_ncbi_fasta_name(hsp.hit_id)]
loc = FeatureLocation(r_offset + hsp.hit_start,
r_offset + hsp.hit_end,
strand=0)
h = gd_record_features.add_feature(SeqFeature(loc),
color=color,
border=border)
gd_diagram.cross_track_links.append(
CrossLink(q, h, color, border, flip))
#Now add the unmatched contigs on outside
position = 0
gd_contig_features = None
unplaced = 0
for contig in SeqIO.parse(assembly_fasta, "fasta"):
contig_id = contig.id
if contig_id in contigs_shown:
continue
#print("Adding unmapped contig %s (len %i bp), offset now %i" % (contig_id, contig_len, position))
unplaced += 1
contig_len = len(contig)
if output_fasta:
if min_len <= contig_len:
fasta_handle.write(contigs.get_raw(contig_id))
gB1.add_feature(i, label=False,\
label_position="start",color=color_atual)
# Marca na figura os trechos sintenicos
for b in blast:
qstart = int(b.split("\t")[0])
qend = int(b.split("\t")[1])
sstart = int(b.split("\t")[2])
send = int(b.split("\t")[3])
identidade = (float(b.split("\t")[4])) / 100
#identidade = (float(b.split("\t")[4])*0.8)/100
# Detectando inversoes
qinv = qend - qstart
sinv = send - sstart
if (qinv > 0 and sinv > 0) or \
(qinv < 0 and sinv < 0):
cor = colors.Color\
(1,.341176,.341176,identidade)
else:
cor = colors.firebrick
if identidade >= 1:
gd.cross_track_links.append(CrossLink((gA, \
qstart, qend),(gB, sstart, send),color=cor))
gd.draw(format="linear", pagesize=(8*cm,29.7*cm), \
fragments=1)
gd.write(name + ".pdf", "PDF")
def draw_alignment(self, _gdd, query_id, query_length, blast_hits):
# draw reference
gdt_features = _gdd.new_track(1,
greytrack=False,
start=1,
end=self.ref_length)
gds_features = gdt_features.new_set()
max_length = 0
for annot in self.annotation:
if annot[-1] == '1':
strand = 1
elif annot[-1] == '-1':
strand = -1
else:
strand = None
start, end = int(annot[0]), int(annot[1])
if end > max_length:
max_length = end
feature = SeqFeature(FeatureLocation(start, end),
strand=strand,
id=annot[2])
gds_features.add_feature(feature,
label=True,
name=annot[2],
sigil='BOX',
label_size=14,
label_angle=0,
arrowhead_length=0.1,
arrowshaft_height=1)
gds_features.add_feature(SeqFeature(
FeatureLocation(self.start, self.start + 1)),
label=True,
name='PCR start',
label_size=14,
color='black')
_gdd.draw(format='linear', fragments=1, start=1, end=max_length)
# draw query sequence
gdt_features = _gdd.new_track(1,
greytrack=True,
start=0,
end=query_length,
name=query_id)
gds_features = gdt_features.new_set()
for r in blast_hits:
feature = SeqFeature(FeatureLocation(r[0], r[1]), strand=r[-1])
gds_features.add_feature(feature, label=False, sigil='BOX')
gdt_features.greytrack_fontcolor = colors.black
gdt_features.greytrack_fontsize = 12
# draw cross link
for r in blast_hits:
alpha = self.scale_color(r[4])
color = colors.linearlyInterpolatedColor(colors.white,
colors.firebrick, 0, 100,
alpha)
feature = SeqFeature(FeatureLocation(r[0], r[1]), strand=r[-1])
gds_features.add_feature(feature, label=False, sigil='BOX')
if max(r[2], r[3]) < self.start:
continue
link_xy = CrossLink((_gdd.tracks[1], r[0], r[1]),
(_gdd.tracks[2], r[2], r[3]), color)
_gdd.cross_track_links.append(link_xy)
_gdd.draw(format='linear', fragments=1)
def plot_multiple_regions_crosslink2(target_protein_list, region_record_list,
plasmid_list, out_name):
gd_diagram = GenomeDiagram.Diagram("geomic_region")
feature_sets = []
max_len = 0
records = dict((rec.name, rec) for rec in region_record_list)
n_records = len(region_record_list)
record_length = [len(record) for record in region_record_list]
for i, record in enumerate(region_record_list):
max_len = max(max_len, len(record))
#print "i", i
#Allocate tracks 3 (top), 1 (bottom) for region 1 and 2
#(empty tracks 2 useful white space to emphasise the cross links
#and also serve to make the tracks vertically more compressed)
gd_track_for_features = gd_diagram.new_track(
(2 * n_records - 1) - 2 * i,
name=record.name,
greytrack=True,
height=0.5,
start=0,
end=len(record))
if record.name not in feature_sets:
feature_sets.append(gd_track_for_features.new_set())
else:
print("already in feature_sets!")
print(record)
quit
for x in range(0, len(region_record_list) - 1):
#print "x", x
features_X = region_record_list[x].features
features_Y = region_record_list[x + 1].features
set_X = feature_sets[x]
set_Y = feature_sets[x + 1]
for feature_1 in features_X:
if feature_1.type != "CDS":
continue
for feature_2 in features_Y:
if feature_2.type != "CDS":
continue
try:
group1 = feature_1.qualifiers["orthogroup"][0]
group2 = feature_2.qualifiers["orthogroup"][0]
except:
group1 = "one_singleton"
group2 = "two_singleton"
if group1 == group2:
border = colors.lightgrey
color = colors.lightgrey
F_x = set_X.add_feature(SeqFeature(
FeatureLocation(feature_1.location.start,
feature_1.location.end,
strand=0)),
color=color,
border=border)
F_y = set_Y.add_feature(SeqFeature(
FeatureLocation(feature_2.location.start,
feature_2.location.end,
strand=0)),
color=color,
border=border)
gd_diagram.cross_track_links.append(
CrossLink(F_x, F_y, color, border))
#for x in range(0,len(region_record_list)-1):
x = 0
for n, record in enumerate(region_record_list):
gd_feature_set = feature_sets[n]
i = 0
if plasmid_list[x]:
#print "PLASMID!!!"
color1 = colors.HexColor('#2837B7')
color2 = colors.blue
else:
color1 = colors.HexColor('#40F13A')
color2 = colors.HexColor('#0F600C')
for feature in record.features:
if feature.type != "CDS":
continue
try:
a = feature.qualifiers["locus_tag"]
except:
# cas des pseudogenes qui sont des CDS mais n'ont pas de protein ID
continue
if len(gd_feature_set) % 2 == 0:
color = color1
else:
color = color2
#try:
# try:
# group = protein_id2group[feature.qualifiers["protein_id"][0]]
# except:
# group = protein_id2group[feature.qualifiers["protein_id"][1]]
#except:
# # no group attributed: singleton => special color
# color = colors.HexColor('#E104C0')
for target_protein in target_protein_list:
if target_protein in feature.qualifiers["locus_tag"]:
#print "target prot!"
color = colors.red
gd_feature_set.add_feature(feature,
sigil="ARROW",
color=color,
label=True,
label_position="middle",
label_strand=1,
label_size=12,
label_angle=45)
i += 1
x += 1
#print "max", max_len
#print "n records", len(region_record_list)
if len(region_record_list) == 2:
hauteur = 700
else:
hauteur = 250 * len(region_record_list)
largeur = max(record_length) / 30
#print "hauteur", hauteur
#print "largeur", largeur
#gd_diagram.set_page_size(, orientation)
if hauteur > largeur:
gd_diagram.draw(format="linear",
pagesize=(hauteur, largeur),
orientation='portrait',
fragments=1,
start=0,
end=max_len)
else:
gd_diagram.draw(format="linear",
pagesize=(hauteur, largeur),
orientation='landscape',
fragments=1,
start=0,
end=max_len)
#print "writing diagram", out_name
gd_diagram.write(out_name, "SVG")
track_x_name = name_for_featureset.name
track_x = name_for_featureset
if name_for_featureset[feature_number].name == cross_link_feature_B and name_for_featureset.name != str(track_x_name):
feature_y = name_for_featureset[feature_number]
track_y_name = name_for_featureset.name
track_y = name_for_featureset
color = colors.linearlyInterpolatedColor(colors.white, colors.firebrick,
0, 100, score)
border = colors.lightgrey
# link_xy = CrossLink((track_x, feature_x.location.start, feature_x.location.end),
# (track_y, feature_y.location.start, feature_y.location.end),
# color, colors.lightgrey)
# gd_diagram.cross_track_links.append(link_xy)
gd_diagram.cross_track_links.append(CrossLink(feature_x, feature_y, color, border))
print('MATCH')
i += 1
#print (len(name_for_featureset))
#for score, nameA, nameB in A_vs_B:
#if name_for_featureset.name == nameA:
# print(score)
# print(nameA)
#print(name_for_featureset.features[4].name)
#for score, x, y in A_vs_B:
#color = colors.linearlyInterpolatedColor(colors.white, colors.firebrick, 0, 100, 50)
# border = colors.lightgrey
def plot_multiple_regions_crosslink(target_protein_list,
region_record_list,
plasmid_list,
out_name,
biodb_name="chlamydia_03_15",
color_locus_list=[],
flip_record_based_on_first=True,
color_orthogroup_list=[]):
import matplotlib.cm as cm
from matplotlib.colors import rgb2hex
import matplotlib as mpl
import MySQLdb
import os
sqlpsw = os.environ['SQLPSW']
norm = mpl.colors.Normalize(vmin=-30, vmax=100)
cmap = cm.Blues
m = cm.ScalarMappable(norm=norm, cmap=cmap)
conn = MySQLdb.connect(
host="127.0.0.1", # your host, usually localhost
user="******", # your username
passwd=sqlpsw, # your password
db="orth_%s" % biodb_name) # name of the data base
cursor = conn.cursor()
gd_diagram = GenomeDiagram.Diagram("geomic_region")
feature_sets = []
max_len = 0
records = dict((rec.name, rec) for rec in region_record_list)
n_records = len(region_record_list)
record_length = [len(record) for record in region_record_list]
if flip_record_based_on_first:
region_record_list_flip = [region_record_list[0]]
region_record_list_flip[0].name = region_record_list_flip[
0].description
for x in range(0, len(region_record_list) - 1):
same_strand_count = 0
different_strand_count = 0
features_X = region_record_list[x].features
features_Y = region_record_list[x + 1].features
for feature_1 in features_X:
if feature_1.type != "CDS":
continue
for feature_2 in features_Y:
if feature_2.type != "CDS":
continue
try:
group1 = feature_1.qualifiers["orthogroup"][0]
group2 = feature_2.qualifiers["orthogroup"][0]
if group1 == group2:
strand1 = feature_1.location.strand
strand2 = feature_2.location.strand
if strand1 == strand2:
same_strand_count += 1
else:
different_strand_count += 1
except:
pass
if different_strand_count > same_strand_count:
region_record_list[x + 1] = region_record_list[
x + 1].reverse_complement(
id=region_record_list[x + 1].id,
name=region_record_list[x + 1].description)
else:
region_record_list[x +
1].name = region_record_list[x +
1].description
#region_record_list = region_record_list_flip
for i, record in enumerate(region_record_list):
max_len = max(max_len, len(record))
#Allocate tracks 3 (top), 1 (bottom) for region 1 and 2
#(empty tracks 2 useful white space to emphasise the cross links
#and also serve to make the tracks vertically more compressed)
gd_track_for_features = gd_diagram.new_track(
(1 * n_records - 1) - 1 * i,
name=record.name,
greytrack=True,
height=0.4,
start=0,
end=len(record))
if record.name not in feature_sets:
feature_sets.append(gd_track_for_features.new_set())
else:
print("already in feature_sets!")
print(record)
quit
#print 'looping....'
for x in range(0, len(region_record_list) - 1):
features_X = region_record_list[x].features
features_Y = region_record_list[x + 1].features
set_X = feature_sets[x]
set_Y = feature_sets[x + 1]
for feature_1 in features_X:
if feature_1.type != "CDS":
continue
for feature_2 in features_Y:
if feature_2.type != "CDS":
continue
try:
group1 = feature_1.qualifiers["orthogroup"][0]
group2 = feature_2.qualifiers["orthogroup"][0]
except:
group1 = "one_singleton"
group2 = "two_singleton"
if group1 == group2:
border = colors.lightgrey
color = colors.lightgrey
try:
identity = orthogroup_identity_db.check_identity(
cursor, feature_1.qualifiers["orthogroup"][0],
feature_1.qualifiers["locus_tag"][0],
feature_2.qualifiers["locus_tag"][0])
except:
identity = 0
print(
"problem with identity table %s and locus %s %s" %
(group1, feature_1.qualifiers["locus_tag"][0],
feature_1.qualifiers["locus_tag"][0]))
color2 = colors.HexColor(
rgb2hex(m.to_rgba(float(identity))))
border2 = colors.HexColor(
rgb2hex(m.to_rgba(float(identity))))
F_x = set_X.add_feature(
SeqFeature(
FeatureLocation(feature_1.location.start,
feature_1.location.end,
strand=0)),
color=color,
border=border,
set_id=feature_1.qualifiers["locus_tag"])
F_y = set_Y.add_feature(SeqFeature(
FeatureLocation(feature_2.location.start,
feature_2.location.end,
strand=0)),
color=color,
border=border)
gd_diagram.cross_track_links.append(
CrossLink(F_x, F_y, color2, border2))
#for x in range(0,len(region_record_list)-1):
x = 0
all_locus = []
for n, record in enumerate(region_record_list):
gd_feature_set = feature_sets[n]
i = 0
if plasmid_list[x]:
#print "PLASMID!!"
color1 = colors.HexColor('#2837B7')
color2 = colors.blue
else:
color1 = colors.HexColor('#40F13A')
color2 = colors.HexColor('#0F600C')
one_row_locus = []
for feature in record.features:
if feature.type == "tblast_target":
feature.name = 'match'
gd_feature_set.add_feature(feature,
sigil="BOX",
color="#ff4a0c86",
label=False,
label_position="middle",
label_size=25,
label_angle=0)
if feature.type == "assembly_gap":
#print "gap", feature
feature.location.strand = None
gd_feature_set.add_feature(feature,
sigil="BOX",
color="red",
label=True,
label_position="middle",
label_strand=1,
label_size=14,
label_angle=40)
if feature.type == "rRNA":
gd_feature_set.add_feature(feature,
sigil="ARROW",
color="orange",
label=True,
label_position="middle",
label_strand=1,
label_size=10,
label_angle=40)
try:
one_row_locus.append(feature.qualifiers["locus_tag"][0])
except:
pass
if feature.type == "tRNA":
gd_feature_set.add_feature(feature,
sigil="ARROW",
color="orange",
label=True,
label_position="middle",
label_strand=1,
label_size=10,
label_angle=40)
try:
one_row_locus.append(feature.qualifiers["locus_tag"][0])
except:
print('no locus tag for:')
print(feature)
if feature.type == "repeat_region":
gd_feature_set.add_feature(feature,
sigil="BOX",
color="blue",
label=True,
label_position="middle",
label_strand=1,
label_size=14,
label_angle=40)
if 'pseudo' in feature.qualifiers:
gd_feature_set.add_feature(feature,
sigil="OCTO",
color="#6E6E6E",
label=True,
label_position="middle",
label_strand=1,
label_size=10,
label_angle=40)
elif feature.type != "CDS":
continue
else:
try:
a = feature.qualifiers["locus_tag"][0]
except:
# cas des pseudogenes qui sont des CDS mais n'ont pas de protein ID
continue
try:
g = feature.qualifiers["orthogroup"][0]
except:
# cas des pseudogenes qui sont des CDS mais n'ont pas de protein ID
continue
if a in color_locus_list:
#print '###########################', a, color_locus_list
if len(gd_feature_set) % 2 == 0:
color = colors.HexColor('#ca4700')
else:
color = colors.HexColor('#fd7a32')
else:
if len(gd_feature_set) % 2 == 0:
color = color1
else:
color = color2
if g in color_orthogroup_list:
#print '###########################', a, color_locus_list
if len(gd_feature_set) % 2 == 0:
color = colors.HexColor('#ca4700')
else:
color = colors.HexColor('#fd7a32')
else:
if len(gd_feature_set) % 2 == 0:
color = color1
else:
color = color2
#try:
# try:
# group = protein_id2group[feature.qualifiers["protein_id"][0]]
# except:
# group = protein_id2group[feature.qualifiers["protein_id"][1]]
#except:
# # no group attributed: singleton => special color
# color = colors.HexColor('#E104C0')
for target_protein in target_protein_list:
if target_protein in feature.qualifiers["locus_tag"]:
#print "target prot!"
color = colors.red
gd_feature_set.add_feature(feature,
sigil="ARROW",
color=color,
label=True,
label_position="middle",
label_strand=1,
label_size=10,
label_angle=40)
i += 1
try:
one_row_locus.append(feature.qualifiers["locus_tag"][0])
except:
print('no locus tag for:')
print(feature)
all_locus = one_row_locus + all_locus
x += 1
#print "max", max_len
#print "n record", len(region_record_list)
if len(region_record_list) == 2:
hauteur = 300
else:
hauteur = 150 * len(region_record_list)
largeur = max(record_length) / 30
#print "hauteur", hauteur
#print "largeur", largeur
#gd_diagram.set_page_size(, orientation)
if hauteur > largeur:
gd_diagram.draw(format="linear",
pagesize=(hauteur, largeur),
orientation='portrait',
fragments=1,
start=0,
end=max_len)
else:
gd_diagram.draw(format="linear",
pagesize=(hauteur, largeur),
orientation='landscape',
fragments=1,
start=0,
end=max_len)
#print "writing diagram", out_name
#gd_diagram.write(out_name, "SVG")
import io
from chlamdb.plots import edit_svg
svg_diagram = io.StringIO()
gd_diagram.write(svg_diagram, "SVG")
svg_diagram.flush()
#gd_diagram
with_links = edit_svg.edit_svg(svg_diagram.getvalue(), all_locus,
biodb_name)
with_links.write(out_name)
png_name = out_name.split('.')[0] + '.png'
#png_handle = open(png_name, 'w')
#gd_diagram.write(png_handle, "PNG")
#png_handle.close()
try:
cmd = 'chmod 444 %s' % out_name
except:
pass
from chlamdb.biosqldb import shell_command
#print cmd
shell_command.shell_command(cmd)
return all_locus
if name_for_featureset[
feature_number].name == cross_link_feature_B and name_for_featureset.name != str(
track_x_name):
features_y_link.insert(number,
name_for_featureset[feature_number])
#feature_y_n = name_for_featureset[feature_number]
track_y_name = name_for_featureset.name
track_y = name_for_featureset
print("Test FEATUREA 2: ")
print(number)
print(name_for_featureset[feature_number].name)
print(features_y_link)
for feature_x, feature_y in zip(features_x_link, features_y_link):
#print("SCORE: " + str(count_feature))
score = A_vs_B[count_feature][0]
color = colors.linearlyInterpolatedColor(colors.white,
colors.firebrick, 0, 100,
score)
border = colors.lightgrey
gd_diagram.cross_track_links.append(
CrossLink(feature_x, feature_y, color, border))
count_feature += 1
i += 1
gd_diagram.draw(format='linear', pagesize='A4', fragments=1, start=0)
gd_diagram.write(diagram_name + ".pdf", "pdf")
gd_diagram.write(diagram_name + ".svg", "SVG")
print fname
color = colors.Color(0.8, 0.8, 0.8, 0.9)
border = colors.lightgrey
with open("ch_comp.out") as f:
for line in f:
ls = line.split()
id1 = ls[0][7:]
if id1 in rnconv:
id1 = rnconv[id1]
id2 = ls[1]
print(id1, id2)
if id1 in rbars1 and id2 in rbars2:
gd_diagram.cross_track_links.append(
CrossLink(rbars1[id1], rbars2[id2], color, border))
for pair in [("rrs", "rrs"), ("rrl", "rrl")]:
if pair[0] in rbars1 and pair[1] in rbars2:
gd_diagram.cross_track_links.append(
CrossLink(rbars1[pair[0]], rbars2[pair[1]], color, border))
print border
gd_diagram.draw(format="linear", pagesize='A4', fragments=1)
#gd_diagram.write(name + ".pdf", "PDF")
gd_diagram.write(name + "_linear.png", "PNG")
gd_diagram.write(name + "_linear.eps", "EPS")
#gd_diagram.write(name + ".svg", "SVG")
sys_exit(
"Could not find offset key %r for hit %r in dict (query id %r)"
% (hack_ncbi_fasta_name(
hsp.hit_id), hsp.hit_id, hsp.query_id))
else:
sys_exit(
"Could not find offset for hit %r in dict (query id %r)"
% (hsp.hit_id, hsp.query_id))
loc = FeatureLocation(r_offset + hsp.hit_start,
r_offset + hsp.hit_end,
strand=0)
hit = gd_ref_features.add_feature(SeqFeature(loc),
color=color,
border=border)
gd_diagram.cross_track_links.append(
CrossLink(query, hit, color, border, flip))
offset += SPACER + contig_len
# Ready for next pairwise comparison,
reference_fasta = assembly_fasta
ref_offsets = contig_offsets
gd_ref_features = gd_contig_features
# Set size based on max track length?
page = (2 * cm + 5 * cm * len(assemblies_fasta), 100 * cm * max_len / 5000000)
gd_diagram.draw(format="linear",
fragments=1,
pagesize=page,
start=0,
end=max_len)
def GenomeMap(file, GenomeId, grid=10000, cross=True):
# print(GenomeId)
gd_diagram = GenomeDiagram.Diagram('phages')
with open(file, 'r') as f:
reader = csv.reader(f)
data = list(reader)
records = []
ref = {}
for Id in GenomeId:
try:
record = SeqIO.read(Id + ".gb", "genbank")
except FileNotFoundError or IOError or ValueError:
hd = Entrez.efetch(db="nucleotide",
id=Id,
rettype='gb',
retmode="text")
record = SeqIO.read(hd, 'genbank')
fw = open(Id + '.gb', 'w')
SeqIO.write(record, fw, 'genbank')
fw.close()
os.getcwd()
for i in SeqIO.parse(Id + ".gb", "genbank"):
ref[Id] = i.annotations['keywords']
records.append(record)
feature_sets = {}
max_len = 0
for i, record in enumerate(records):
max_len = max(max_len, len(record))
gd_track_for_features = gd_diagram.new_track(
5 - 2 * i,
name=record.description,
greytrack=True,
greytrack_fontsize=16,
greytrack_labels=1,
largetick=True,
smalltick=True,
scale_ticks=True,
scale_largeticks=0.5,
scale_smallticks=0.1,
scale_largetick_interval=grid,
scale_smalltick_interval=grid / 20,
scale_largetick_labels=True,
start=0,
end=len(record),
)
assert record.name not in feature_sets
feature_sets[record.id] = gd_track_for_features.new_set()
for crosslink in data:
if not cross:
break
set_X = feature_sets[crosslink[0].split(' ')[0]]
set_Y = feature_sets[crosslink[1].split(' ')[0]]
# 手动划分连接类型时使用
# score = 100
# try:
# if crosslink[7] == 1 or crosslink[7] == -1:
# score = 100
# except TypeError:
# score = 50
if crosslink[0].split(' ')[0] in CLASS1 and crosslink[1].split(
' ')[0] in CLASS1:
color = colors.linearlyInterpolatedColor(
colors.green, colors.yellow, 0, len(GenomeId),
GenomeId.index(crosslink[1].split(' ')[0]))
elif crosslink[0].split(' ')[0] in CLASS2 and crosslink[1].split(
' ')[0] in CLASS2:
color = colors.linearlyInterpolatedColor(
colors.purple, colors.red, 0, len(GenomeId),
GenomeId.index(crosslink[1].split(' ')[0]))
else:
color = colors.linearlyInterpolatedColor(
colors.blue, colors.cyan, 0, len(GenomeId),
GenomeId.index(crosslink[1].split(' ')[0]))
# color = list(colors.getAllNamedColors().keys())[GenomeId.index(crosslink[1].split(' ')[0]) * 17 + 17 % 163]
F_x = set_X.add_feature(
SeqFeature(
FeatureLocation(int(crosslink[2]), int(crosslink[3]),
strand=0)),
color=color,
border=color,
)
F_y = set_Y.add_feature(
SeqFeature(
FeatureLocation(int(crosslink[4]), int(crosslink[5]),
strand=0)),
color=color,
border=color,
)
link_xy = CrossLink(F_x, F_y, color, color)
gd_diagram.cross_track_links.append(link_xy)
for record in records:
gd_feature_set = feature_sets[record.id]
# 矫正ori
for feature in record.features:
if feature.type == 'rep_origin':
print(record.description + ' 的起始位点在:' +
str(feature.location.start))
record = record[feature.location.
start:] + record[:feature.location.start]
if record.features[0].strand == -1:
print('daole')
record = record.reverse_complement(id=True,
name=True,
description=True,
features=True,
annotations=True,
letter_annotations=True)
break
# 务必绘制反向互补序列时手动开启
# record = record.reverse_complement(id=True, name=True, description=True, features=True,
# annotations=True, letter_annotations=True)
print(record.description + ' 的起始位点已校正')
# 画features
i = 0
if ref[record.id] != ['']:
for feature in record.features:
if feature.type != "gene":
continue
color = list(colors.getAllNamedColors().keys())[len(feature) %
163]
gd_feature_set.add_feature(feature,
color=color,
label=True,
label_size=10,
label_angle=90,
sigil="ARROW",
arrowshaft_height=1.0,
arrowhead_length=0.1)
i += 1
elif ref[record.id] == ['']:
for feature in record.features:
if feature.type != "CDS":
continue
color = list(colors.getAllNamedColors().keys())[len(feature) %
163]
gd_feature_set.add_feature(feature,
color=color,
label=True,
label_size=10,
label_angle=90,
sigil="ARROW",
arrowshaft_height=1.0,
arrowhead_length=0.2)
i += 1
# 用来手动添加重组位点
# for pos in recombinations:
# if pos in record.features:
# gd_feature_set.add_feature(feature, color=color, label=True, label_size=10, label_angle=90,
# sigil="ARROW", arrowshaft_height=1.0, arrowhead_length=0.1)
if not cross:
# 用来绘制单一序列
gd_diagram.draw(format="linear",
pagesize='A4',
fragments=5,
start=0,
end=max_len,
fragment_size=1)
gd_diagram.write("T7.pdf", "PDF")
else:
# 用来绘制比对序列
gd_diagram.draw(format="linear",
pagesize=(10 * len(GenomeId) * cm, 120 * cm),
fragments=1,
start=0,
end=max_len,
fragment_size=1)
gd_diagram.write(output, "PDF")
print("已输出为PDF")
set_Y = feature_sets[Y]
for score, x, y in X_vs_Y:
color = colors.linearlyInterpolatedColor(colors.white, colors.firebrick,
0, 100, score)
border = colors.lightgrey
f_x = get_feature(features_X, x)
F_x = set_X.add_feature(SeqFeature(FeatureLocation(f_x.location.start,
f_x.location.end,
strand=0)),
color=color, border=border)
f_y = get_feature(features_Y, y)
F_y = set_Y.add_feature(SeqFeature(FeatureLocation(f_y.location.start,
f_y.location.end,
strand=0)),
color=color, border=border)
gd_diagram.cross_track_links.append(CrossLink(F_x, F_y, color, border))
for record, gene_colors in zip([A_rec, B_rec, C_rec],
[A_colors, B_colors, C_colors]):
gd_feature_set = feature_sets[record.name]
i = 0
for feature in record.features:
if feature.type != "gene":
# Exclude this feature
continue
try:
g_color = gene_colors[i]
except IndexError:
print("Don't have color for %s gene %i" % (record.name, i))
identity = float(each_line_split[2])
query_start = int(each_line_split[6])
query_end = int(each_line_split[7])
target_start = int(each_line_split[8])
target_end = int(each_line_split[9])
# use color to reflect identity
color = colors.linearlyInterpolatedColor(colors.white, colors.red, 50,
100, identity)
# determine which is which (query/target to contig_1/contig_2)
# if query is contig_1
if query == gene1_contig.name:
link = CrossLink(
(contig_1_gene_content_track, query_start, query_end),
(contig_2_gene_content_track, target_start, target_end),
color=color,
border=color,
flip=False)
diagram.cross_track_links.append(link)
# if query is contig_2
elif query == gene2_contig.name:
link = CrossLink(
(contig_2_gene_content_track, query_start, query_end),
(contig_1_gene_content_track, target_start, target_end),
color=color,
border=color,
flip=False)
diagram.cross_track_links.append(link)
# Draw and Export
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')
if i.type == "rRNA":
color_atual = colors.blue
gB1.add_feature(i,
label=False,
label_position="start",
color=color_atual)
# Marca na figura os trechos sintenicos
for b in blast:
qstart = int(b.split("\t")[0])
qend = int(b.split("\t")[1])
sstart = int(b.split("\t")[2])
send = int(b.split("\t")[3])
identidade = (float(b.split("\t")[4]) * 0.8) / 100
# Detectando inversoes
qinv = qend - qstart
sinv = send - sstart
if (qinv > 0 and sinv > 0) or (qinv < 0 and sinv < 0):
cor = colors.Color(1, .341176, .341176, identidade)
else:
cor = colors.firebrick
gd.cross_track_links.append(
CrossLink((gA, qstart, qend), (gB, sstart, send), color=cor))
gd.draw(format="linear", pagesize=(8 * cm, 29.7 * cm), fragments=1)
gd.write(name + ".pdf", "PDF")
