else:
start = int(sp_line[2])
end = int(sp_line[3])
strand = 1
if series in series_features:
series_features[series].append((gene_name, strand, start, end))
else:
series_features[series] = [(gene_name, strand, start, end)]
series_indexes[series] = series_index
series_index = series_index + 1
start = sys.maxsize
end = -1
gdd = GenomeDiagram.Diagram("diagram", tracklines=False, y=0.4)
gd_track_for_features = gdd.new_track(1,
scale=True,
height=1,
scale_smallticks=0)
gds_features = gd_track_for_features.new_set()
seed(int(arg.random_seed))
for series in series_features.keys():
if series_indexes[series] < len(colors_list):
current_color = colors_list[series_indexes[series]]
else:
current_color = colors.Color(random(), random(), random())
for i in range(0, len(series_features[series])):
'''
first try, not using Diagram_class
'''
from Bio.SeqFeature import SeqFeature, FeatureLocation
from Bio.Graphics import GenomeDiagram
from Bio.Graphics.GenomeDiagram import CrossLink
from reportlab.lib.units import cm
from reportlab.lib import colors
gdd = GenomeDiagram.Diagram('Diagram')
gdt1_features = gdd.new_track(1, greytrack=False)
gds1_features = gdt1_features.new_set()
gdt2_features = gdd.new_track(1, greytrack=False)
gds2_features = gdt2_features.new_set()
inFile = open('ABL1_NC')
NC_len = 5894
num = 0
startend = []
for line in inFile:
num += 1
color = colors.linearlyInterpolatedColor(colors.white, colors.firebrick, 0,
10, num)
line = line.strip()
fields = line.split('\t')
q_start = int(fields[7])
q_end = int(fields[8])
s_start = int(fields[9])
s_end = int(fields[10])
#!/home/pjsola/env/bin/python
import os
import csv
from Bio.SeqFeature import SeqFeature, FeatureLocation
from reportlab.lib import colors
from reportlab.lib.units import cm
from Bio.Graphics import GenomeDiagram
from Bio.Graphics.GenomeDiagram import CrossLink
diagram_name = 'TEST_CL'
gd_diagram = GenomeDiagram.Diagram(diagram_name)
dict_records = {'NC_016838':122799, 'NC_016839':105974, 'NC_016846':111195}
#NC_016838.1 vs NC_016839.1 made up reltions
A_vs_B = [
(99, "mcpQ", "tetR"),
(33, "ligA", "rhsC")
]
B_vs_C = [
(99, "tetA", "pld"),
(33, "rhsC", "traC")
]
i = 0
for record,record_length in dict_records.items():
# Allocate tracks 5 (top), 3, 1 (bottom) for A, B, C
# (empty tracks 2 and 4 add useful white space to emphasise the cross links
# and also serve to make the tracks vertically more compressed)
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
from reportlab.lib import colors
from reportlab.lib.units import cm
from Bio.Graphics import GenomeDiagram
from Bio import SeqIO
from Bio.SeqFeature import SeqFeature, FeatureLocation
import sys
record = next(SeqIO.parse(sys.argv[1], "genbank"))
print(record)
gd_diagram = GenomeDiagram.Diagram(record.id)
gd_track_for_features = gd_diagram.new_track(1, name="Annotated Features")
gd_feature_set = gd_track_for_features.new_set()
contig = SeqFeature(FeatureLocation(0, len(record.seq)))
gd_feature_set.add_feature(contig,
sigil="ARROW",
color="black",
label=True,
name="1",
arrowshaft_height=1.0,
label_size=14,
label_angle=0)
gd_feature_set.add_feature(contig,
sigil="ARROW",
color="black",
label=True,
name="1",
arrowshaft_height=1.0,
label_size=14,
label_angle=0)
for feature in record.features:
if feature.type != "gene":
# import library to create genome diagram
from Bio import SeqIO
from Bio.Graphics import GenomeDiagram
# import library to parse file
from reportlab.lib import colors
from reportlab.lib.units import cm
# read genome file
record = SeqIO.read("Genome.gb", "genbank")
# make diagram for the genome
gd_diagram = GenomeDiagram.Diagram("DNA sequence visualization")
gd_track_for_features = gd_diagram.new_track(1, name="Annotated Features")
gd_feature_set = gd_track_for_features.new_set()
# add features to diagram
for feature in record.features:
if feature.type != "gene":
continue
if len(gd_feature_set) % 2 == 0:
#add colors to diagram
color = colors.green
else:
color = colors.darkcyan
gd_feature_set.add_feature(feature,
color=color,
label=True,
label_size=20,
label_color=color)
line = line.split()
rev_data.append((int(line[1]), float(line[2])))
if float(line[2]) > scale_height:
scale_height = float(line[2])
rev_data.append((len(record.seq) + 5, scale_height))
print(filename)
print("Max height: " + str(scale_height))
if "ICP1" in filename:
scale_height = 10
elif "chromosome" in filename:
scale_height = 1000
else:
scale_height = 50000
gd_diagram = GenomeDiagram.Diagram("temp")
gd_track_for_features = gd_diagram.new_track(
5,
name="Annotated Features",
height=1,
scale_ticks=0,
greytrack=1,
greytrack_labels=0,
scale=0)
gd_feature_set = gd_track_for_features.new_set(type='feature')
gd_track_for_feature_names = gd_diagram.new_track(
2,
name="Annotated names",
height=1,
scale_ticks=1,
ret_list.append((i, x))
i = i + 100
return ret_list
graphdata2 = gcSkewData(SeqUtils.GC_skew(seq_string))
gdgs2 = GenomeDiagram.GraphSet('GC Skew')
gdgs2.new_graph(graphdata, 'GC Skew', style='line', linewidth=2)
gdt3 = GenomeDiagram.Track('GC Skew', greytrack=1, greytrack_labels=4)
gdt3.add_set(gdgs2)
track_list.append(gdt3)
gd_diagram = GenomeDiagram.Diagram("Tomato Curly Stunt Virus, complete genome",
track_size=0.7)
i = 1
for track in track_list:
gd_diagram.add_track(track, i)
i = i + 1
gd_diagram.draw(format="circular",
circular=True,
pagesize=(50 * cm, 50 * cm),
start=0,
end=len(record),
circle_core=0.3)
gd_diagram.write("circularGenomeTCSV.pdf", "PDF", dpi=72)
def record2graph(fnames, beds, r, minlog, window, verbose):
""" """
#create diagram
gdd = GenomeDiagram.Diagram() #GDDiagram(gb)
#add annotation
gdt1 = gdd.new_track(1,
greytrack=1,
name="Genes and GC",
height=1.5,
scale_smalltick_interval=5 * 10**4,
scale_smallticks=0.15,
scale_largeticks=1.0,
scale_largetick_labels=1,
scale_largetick_interval=250 * 10**3,
scale_fontangle=0)
gdt1.greytrack_fontcolor = colors.black
gdfs = gdt1.new_set("feature")
for feature in r.features:
if feature.type == "CDS":
gdfs.add_feature(feature, color=colors.grey)
#add GC
gdgs = gdt1.new_set("graph")
#get gc graph
gcgraph = seq2gcgraph(r, beds[0][0][0])
gdgs.new_graph(gcgraph,
"GC content",
style="line",
color=colors.blue,
center=50)
#add coverage tracks for each bed tracks file
ggraphs, gtracks = [], []
for i, fn in enumerate(fnames):
#add individual track
gdt = gdd.new_track(i + 2,
greytrack=1,
name=fn,
height=1,
scale_smalltick_interval=5 * 10**4,
scale_smallticks=0.15,
scale_largetick_labels=0,
scale_fontangle=0)
gdt.greytrack_fontcolor = colors.black
#adjust font
fsize = 8 - len(fnames) / 5
if fsize < 1:
fsize = 1
gdt.greytrack_fontsize = fsize
#add feature and graph
gtracks.append(gdt.new_set("feature"))
ggraphs.append(gdt.new_set("graph"))
j = 0
#colorstuple = ("blue","darkgrey","orange")
#first add coverage track
i = 0
for bed, expcount, bed1, expcount1, bed2, expcount2 in zip(
beds[0][0], beds[0][1], beds[1][0], beds[1][1], beds[2][0],
beds[2][1]):
#add SNPs colored data
fdata = bed2SeqFeature(bed1, expcount1, bed2, expcount2, window)
for feature, color in fdata:
gtracks[i].add_feature(feature, colour=color)
#add coverage track
gdata = bed2graph(bed, window, expcount, minlog)
gdgg = ggraphs[i].new_graph(gdata,
fn,
style="line",
linewidth=0.5,
center=0,
color="blue")
i += 1
#write
xl = xr = 0.05
width = 841.8897637795275
height = 595.275590551181
'''if clen<10.0**6:
xr = 1.0 - clen / 10.0**6 * 0.95
else:
width = clen * width / 10.0**6
if len(fnames)>12:
height = len(fnames)/12.0 * height'''
#draw
gdd.draw(format="linear",
pagesize=(width, height),
xl=xl,
xr=xr,
orientation="landscape",
tracklines=0,
fragments=1,
circular=0,
track_size=0.75) # ,pagesize="A3"
return gdd
def cre(project_data_dir="",
global_data_dir="",
feature_name="",
construct_name="CRE",
extract_from="",
main_record_file="",
align_with=[],
primer_list=[],
restriction_interval=[],
rb=[],
write=False,
pagesize="A4",
scale_fontsize=3,
label_size=2,
greytrack_fontsize=7,
x=0.05,
y=0.01,
track_size=0.3):
#define color palettes:
primer_colors = [
colors.orchid, colors.cornflower, colors.lightseagreen, colors.salmon
]
if extract_from:
main_record, main_record_file = extract_feature(
sequence_id=extract_from,
data_dir=global_data_dir,
feature_names=feature_name,
write_file=True)
elif main_record_file:
extract_from = splitext(basename(main_record_file))[0]
if splitext(basename(main_record_file))[1] == ".fasta":
main_record = SeqIO.read(main_record_file, "fasta")
elif splitext(
basename(main_record_file))[1] in [".gb", ".gbk", ".genbank"]:
main_record = SeqIO.read(main_record_file, "gb")
main_record_file = convert_seq(main_record_file, "genbank",
"fasta")
print main_record_file
gdd = GenomeDiagram.Diagram(construct_name + ' Construct Diagram',
x=x,
y=y,
track_size=track_size)
genbank_track, genbank_features = new_track(
gdd,
construct_name + " features",
smalltick=10,
scale_fontsize=scale_fontsize,
greytrack_fontsize=greytrack_fontsize)
for feature in main_record.features:
if "Cre" in str(feature.qualifiers):
color = colors.lavender
else:
color = colors.grey
genbank_features.add_feature(feature,
sigil="ARROW",
color=color,
label_color=color,
label=True,
label_size=label_size,
label_angle=30,
arrowshaft_height=1)
if restriction_interval or rb:
restriction_dict = enzyme_selector(sequence=main_record,
restriction_interval=[0, 690],
genome_frequency=[700, 2000],
deterministic_overhangs=True,
rb=["XceI", "PsuI"])
restriction_track, restriction_features = new_track(
gdd,
construct_name + " restriction sites",
smalltick=10,
scale_fontsize=scale_fontsize,
greytrack_fontsize=greytrack_fontsize)
draw_digest(restriction_features, restriction_dict)
# plotting primers
if primer_list:
primer_colors = cycle(primer_colors)
primer_track, primer_features = new_track(
gdd,
construct_name + " primers",
smalltick=10,
scale_fontsize=scale_fontsize,
greytrack_fontsize=greytrack_fontsize)
for primer_entry in primer_list:
primer_color = primer_colors.next()
add_to_track(primer_features,
global_data_dir + "primers/" + primer_entry[0] +
".fasta",
main_record_file,
annotation=primer_entry[0],
feature_color=primer_color,
label_angle=30,
label_size=label_size)
add_to_track(primer_features,
global_data_dir + "primers/" + primer_entry[1] +
".fasta",
main_record_file,
annotation=primer_entry[1],
feature_color=primer_color,
label_angle=30,
label_size=label_size)
# turn entry names into actual file paths
align_with = [project_data_dir + entry + ".fasta" for entry in align_with]
for i in align_with:
hit_track_back, hit_features_back = new_track(
gdd,
splitext(i)[0][-6:],
smalltick=10,
end=len(SeqIO.read(i, 'fasta')),
scale_fontsize=scale_fontsize,
greytrack_fontsize=greytrack_fontsize)
add_to_track(hit_features_back,
main_record_file,
i,
annotation=" " + construct_name,
feature_color=colors.red,
label_angle=30,
forceone=True,
label_size=label_size)
hit_track, hit_features = new_track(
gdd,
construct_name + " alignment hits",
smalltick=10,
scale_fontsize=scale_fontsize,
greytrack_fontsize=greytrack_fontsize)
hsp_list = add_to_track(hit_features,
i,
main_record_file,
annotation=" " + splitext(i)[0][-6:],
feature_color=colors.red,
label_angle=30,
forceone=True,
label_size=label_size)
record = SeqIO.read(i, "fasta")
#for loop only takes the first alignment, then breaks
for hsp in hsp_list:
truncated_record = SeqRecord(
record.seq[hsp.query_end:],
id="Region downstream of Cre in ePet-cre mice, i=" + i)
write_seq(sequence_write_path=project_data_dir,
record=truncated_record,
ID=splitext(basename(i))[0] + "_3-unmatched")
break
if align_with:
record = SeqRecord(main_record.seq + record.seq[hsp.query_end:],
id="Cre and following bases in ePet-cre construct.")
write_seq(sequence_write_path=project_data_dir,
record=record,
ID="cre-ff-current")
gdd.draw(format="linear",
pagesize=pagesize,
fragments=1,
start=0,
end=len(main_record))
if write:
gdd.write(
"/home/chymera/src/AutoTransGeno/output/" + construct_name +
"_from_" + extract_from + ".pdf", "PDF")
print "/home/chymera/src/AutoTransGeno/output/" + construct_name + "_from_" + extract_from + ".pdf"
return gdd
def geneclusterview(gene_cluster, cutoff):
cursor = asp_con(path='192.38.13.196', user='******', pw='1234')
#recA = SeqIO.read("GC_11_382281_6.gb", "gb")
#recB = SeqIO.read("GC_28_10278_8.gb", "gb")
#recC = SeqIO.read("GC_28_3654_7.gb", "gb")
red = PCMYKColor(0, 100.0, 0.0, 0.0)
blue = PCMYKColor(91.0, 43.0, 0.0, 0.0)
black = PCMYKColor(0, 0, 20.0, 0)
pal = fade(blue,[100,60,40,20,5,0]) # TODO better coloring!
#records = [recA, recB, recC] # for testing
# TODO if isfile
records = seq_builder(gene_cluster, cutoff, save = False)
# records = [recA, recB, recC]
##########################################
# Doing some black magic on rank numbers #
##########################################
hits = []
tagcol = []
#all_h_tags=[]
rank_col = []
for i in records[1:]: # Leaving query cluster out # Exchange hardcoded query cluster later!
tags = []
for n in i.features:
if n.qualifiers['locus_tag'][2:] is None:
print "cannot get rank for this protein"
print n
else:
tags.append(n.qualifiers['locus_tag'][2:]) # Hod to put a 0 here before... don't know why it doesn't need that now...
tags_formatted = "("+str(tags)[1:-1]+")"
#all_h_tags.append(str(tags)[1:-1])
query ="SELECT h_seqkey,CAST(pident AS UNSIGNED) from t_antismash2blast_reduced where clust_id = '%s' and h_seqkey IN %s;" % (gene_cluster,tags_formatted)
cursor.execute(query)
result = list(cursor.fetchall())
query =" SELECT ta.q_seqkey, tx.h_seqkey from t_antismash2blast_reduced as ta left join (SELECT * from t_antismash2blast_reduced as tb where tb.clust_id = '%s' and tb.h_seqkey IN %s ) tx on ta.q_seqkey = tx.q_seqkey where ta.clust_id = '%s' and ta.h_seqkey = ta.q_seqkey; "% (gene_cluster, tags_formatted, gene_cluster)
cursor.execute(query)
ranks_raw = cursor.fetchall()
ranks = {}
rank_counter = 1
for line in ranks_raw:
a,b = line
if b in ranks:
if b!='NULL':
ranks[b].append(rank_counter)
rank_counter+=1
else:
if b!='NULL':
ranks[b] = list()
ranks[b].append(rank_counter)
rank_counter+=1
rank_col.append(ranks)
tag_pid = {}
for line in result:
a,b = line
if a in tag_pid:
tag_pid[a].append(int(b))
else:
tag_pid[a] = list()
tag_pid[a].append(int(b))
hits.append(tag_pid)
tagcol.append(tags)
#print rank_col
#f_all_h_tags= ','.join(all_h_tags)
#formatted = "('"+str(f_all_h_tags)[1:-1]+"')"
#print tagcol
#print '\n'
#print hits
if gene_cluster[-2]=='_':
fix = range(int(gene_cluster[-1])+1)
else:
fix = range(int(gene_cluster[-2:])+1)
print fix
rankA=fix[1:] #HERE!!!!
print rankA
# Let's see if [[] for i in range(len(records))] works because some records from org 3 show empty features
ranklist = [[] for i in range(len(records))]
#print range(len(records))
cols=[[] for i in range(len(records))]
counter= -1
for i in tagcol:
counter+=1
for n in i:
#print counter
if n in rank_col[counter]:
ranklist[counter].append(rank_col[counter][n][0])
else:
ranklist[counter].append(0)
if n in hits[counter]:
calc = round(hits[counter][n][0])
if calc > 80:
index=0
elif calc >60:
index = 1
elif calc > 40:
index = 2
elif calc > 20:
index = 3
else:
index = 4
cols[counter].append(pal[index])
else:
cols[counter].append(black)
#print ranklist
if gene_cluster[-2]=='_':
dummy_int = int(gene_cluster[-1])+1
else:
dummy_int = int(gene_cluster[-2:])+1
colA=[blue]*dummy_int
def set_color():
pass
name = "/home/seth/300asp/Scripts/genecluster_plots/gc_view_%s" % gene_cluster
gd_diagram = GenomeDiagram.Diagram(name)
max_len = 0
#print type([rankA]+ranklist)
#print [rankA]+ranklist
for record, gene_colors, rank in zip(records, [colA]+cols, [rankA]+ranklist):
max_len = max(max_len, len(record))
gd_track_for_features = gd_diagram.new_track(1,
name=record.description + record.id,
greytrack=True,
greytrack_labels = 1,
greytrack_font_rotation = 0,
greytrack_font_colour = Color(1,0,0),
greytrack_fontsize = 5,
axis_labels = True,
scale_smallticks = 0.6,
start=0, end=len(record))
gd_feature_set = gd_track_for_features.new_set()
i = 0
#print len(rank)
#print len(record.features)
#print rank
for feature, single_rank in zip(record.features, rank):
#if feature.type != "gene": # Doesn't work with my records because they don't have genes, only cds.... should work but doesn't....
#Exclude this feature
# continue
if feature.strand == -1:
temp_angle = 180
else:
temp_angle = 1
if single_rank == 0:
temp_name = ''
else:
temp_name = str(single_rank) # str(feature.qualifiers['locus_tag'])+'M'+str(single_rank)
gd_feature_set.add_feature(feature, sigil="BIGARROW",
color=gene_colors[i], label=True,
name = temp_name, # this is too much, maybe implement if other label is found: str(feature.qualifiers['locus_tag'])+'M'+str(single_rank),
label_position="middle",
label_size = 6, label_angle= temp_angle)
i+=1
gd_diagram.draw(format="linear", pagesize='A4', fragments=1, orientation = 'portrait', track_size = 0.6, xl = 0.15,
start=0, end=max_len)
gd_diagram.write(name + ".pdf", "PDF")
#gd_diagram.write(name + ".eps", "EPS")
gd_diagram.write(name + ".svg", "SVG")
def plot_motif_sites(self, cluster_num, motif_num):
"""THIS NEEDS MORE WORK but has the beginnings of something...
TODO: multiple motifs on same tracks, include ALL genes (i.e. in operons that were not included),
do reverse-complement positioning correctly (based on gene strand),
use MAST scan output (from b.tables['motif_annotations'])
"""
from Bio.SeqFeature import SeqFeature, FeatureLocation
from Bio.Graphics import GenomeDiagram
from reportlab.lib.units import cm
from reportlab.lib import colors
"""To get this to work: download http://www.reportlab.com/ftp/fonts/pfbfer.zip
and unzip it into /usr/lib/python2.7/dist-packages/reportlab/fonts/
"""
motif_sites = self.get_motif_sites(cluster_num, motif_num)
pv_range = np.max(
-np.log10(motif_sites.pvalue.values)
) - 4 ## divide -log10(pval) by this to get alpha to use
len_range = np.max(motif_sites.start.values) + 10
gdd = GenomeDiagram.Diagram('Motif sites: %d, %d' %
(cluster_num, motif_num))
for i in range(motif_sites.shape[0]):
gdt_features = gdd.new_track(1,
start=0,
end=len_range,
greytrack=True,
greytrack_labels=1,
name=motif_sites.names.values[i],
scale=True,
greytrack_fontsize=4)
gds_features = gdt_features.new_set()
col = colors.red.clone()
col.alpha = (-np.log10(motif_sites.pvalue.values[i]) -
4) / pv_range
m_start = motif_sites.start.values[i]
m_len = len(motif_sites.seq.values[i])
m_strand = motif_sites.reverse.values[i]
if m_strand == 0:
m_strand = -1
feature = SeqFeature(FeatureLocation(m_start, m_start + m_len - 1),
strand=m_strand)
gds_features.add_feature(feature,
name=str(i + 1),
label=False,
color=col)
gdd.draw(format='linear',
pagesize=(15 * cm, motif_sites.shape[0] * cm / 2),
fragments=1,
start=0,
end=len_range + 10)
##gdd.write("GD_labels_default.pdf", "pdf") ## looks like only output is to file, so do this:
#output = cStringIO.StringIO()
#gdd.write(output, 'png', dpi=300)
#output.seek(0)
output = gdd.write_to_string(output='png', dpi=300)
output = cStringIO.StringIO(output)
img = mpimg.imread(output)
plt.axis('off')
imgplot = plt.imshow(img, interpolation='bicubic')
output.close()
return gdd
def __init__(self):
self.tracks = []
self.gdd = GenomeDiagram.Diagram('Diagram') #Is this name useful ?
max_len += SPACER + len(record)
max_len -= SPACER
if os.path.isfile(reference_genbank):
reference_parser = SeqIO.parse(reference_genbank, "genbank")
else:
reference_parser = SeqIO.parse(reference_fasta, "fasta")
if output_fasta:
sys.stderr.write(
"WARNING - Consider using order_assembly.py instead for FASTA output\n"
)
fasta_handle = open(output_fasta, "w")
fasta_saved_count = 0
fasta_short_dropped = 0
gd_diagram = GenomeDiagram.Diagram("Comparison")
gd_track_for_features = gd_diagram.new_track(1,
name="reference",
greytrack=False,
height=0.5,
start=0,
end=max_len)
gd_feature_set = gd_track_for_features.new_set()
# Add a dark grey background
gd_feature_set.add_feature(SeqFeature(FeatureLocation(0, len(record))),
sigil="BOX",
color="grey",
label=False),
offset = 0
ref_offsets = dict()
def record2graph(fnames, beds, r, minlog, window, verbose):
""" """
#create diagram
gdd = GenomeDiagram.Diagram() #GDDiagram(gb)
#gdd.name = r.id
#add annotation
gdt1 = gdd.new_track(1,
greytrack=1,
name="[%s] Genes and GC" % r.id,
height=1.5,
scale_smalltick_interval=5 * 10**4,
scale_smallticks=0.15,
scale_largeticks=1.0,
scale_largetick_labels=1,
scale_largetick_interval=250 * 10**3,
scale_fontangle=0)
gdt1.greytrack_fontcolor = colors.black
gdfs = gdt1.new_set("feature")
for feature in r.features:
if feature.type == "CDS":
gdfs.add_feature(feature, color=colors.grey)
#add GC
gdgs = gdt1.new_set("graph")
#get gc graph
gcgraph = seq2gcgraph(r, beds[0][0][0])
gdgs.new_graph(gcgraph,
"GC content",
style="line",
color=colors.blue,
center=50)
#add coverage tracks for each bed tracks file
gdgslist = []
for i, fn in enumerate(fnames):
#add individual track
gdt = gdd.new_track(i + 2,
greytrack=1,
name=fn,
height=2,
scale_smalltick_interval=5 * 10**4,
scale_smallticks=0.15,
scale_largetick_labels=0,
scale_fontangle=0)
gdt.greytrack_fontcolor = colors.black
gdgslist.append(gdt.new_set("graph"))
colorstuple = ["blue", "darkgrey", "orange"]
colorstuple.reverse()
beds.reverse()
for j, bedexps in enumerate(beds):
for i, (bed, expcount) in enumerate(zip(bedexps[0], bedexps[1])):
if bed:
gdata = bed2graph(bed, window, expcount, minlog)
#add graph to track
linewidth = 0.3
if j == 2:
linewidth = 1.0
gdgg = gdgslist[i].new_graph(gdata,
fn,
style="line",
linewidth=linewidth,
center=0,
color=colorstuple[j])
clen = gdgslist[i].range()[1]
#write
xl = xr = 0.05
width = 841.8897637795275
height = 595.275590551181
'''if clen<10.0**6:
xr = 1.0 - clen / 10.0**6 * 0.95
else:
width = clen * width / 10.0**6
if len(fnames)>12:
height = len(fnames)/12.0 * height'''
#draw
gdd.draw(format="linear",
pagesize=(width, height),
xl=xl,
xr=xr,
orientation="landscape",
tracklines=0,
fragments=1,
circular=0,
track_size=0.75) # ,pagesize="A3"
return gdd
#!/home/pjsola/env/bin/python
import os
import csv
from Bio.SeqFeature import SeqFeature, FeatureLocation
from reportlab.lib import colors
from reportlab.lib.units import cm
from Bio.Graphics import GenomeDiagram
diagram_name = 'TEST_3'
gdd = GenomeDiagram.Diagram(diagram_name)
dict_records = {
'NC_016838.1': 122799,
'NC_016839.1': 105974,
'NC_016846.1': 111195
}
for record, record_length in dict_records.items():
gd_track_for_features = gdd.new_track(1,
name=record,
greytrack=True,
start=0,
end=record_length)
gd_set_features = gd_track_for_features.new_set()
with open('KPN.gff.forward.coordinates', 'r') as bed_forward_file:
bed_readed = csv.reader(bed_forward_file, delimiter="\t")
#record = None
(28, "orf54", "lin2566"),
]
def get_feature(features, id, tags=("locus_tag", "gene", "old_locus_tag")):
"""Search list of SeqFeature objects for an identifier under the given tags."""
for f in features:
for key in tags:
# tag may not be present in this feature
for x in f.qualifiers.get(key, []):
if x == id:
return f
raise KeyError(id)
gd_diagram = GenomeDiagram.Diagram(name)
feature_sets = {}
max_len = 0
for i, record in enumerate([A_rec, B_rec, C_rec]):
max_len = max(max_len, len(record))
# Allocate tracks 5 (top), 3, 1 (bottom) for A, B, C
# (empty tracks 2 and 4 add 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(5 - 2 * i,
name=record.name,
greytrack=True,
height=0.5,
start=0,
end=len(record))
assert record.name not in feature_sets
feature_sets[record.name] = gd_track_for_features.new_set()
def write_schemadelica_plot(self):
"""Write schemadelica plot as SVG and PDF."""
gd_diagram = GenomeDiagram.Diagram("Primer Scheme", track_size=0.15)
primer_feature_set = GenomeDiagram.FeatureSet()
# make the gc track
window = 50
gc_set = GenomeDiagram.GraphSet("GC content")
graphdata1 = self.apply_to_window(self.primary_ref.seq, window,
self.calc_gc)
gc_set.new_graph(
graphdata1,
"GC content",
style="line",
color=colors.violet,
altcolor=colors.purple,
)
gc_track = GenomeDiagram.Track("GC content",
height=1.5,
greytrack=0,
scale_largetick_interval=1e3)
gc_track.add_set(gc_set)
# make the primer track
for r in self.regions:
region = str(r.region_num)
strand = 1 if r.region_num % 2 else -1
fwd_feature = SeqFeature(
FeatureLocation(r.left.start, r.left.end, strand=strand))
rev_feature = SeqFeature(
FeatureLocation(r.right.end, r.right.start, strand=strand))
region_feature = SeqFeature(
FeatureLocation(r.left.start, r.right.start, strand=strand))
primer_color = colors.red
region_color = colors.palevioletred
primer_feature_set.add_feature(
region_feature,
color=region_color,
name=region,
label=True,
label_position="middle",
label_angle=0 if strand == 1 else -180,
)
primer_feature_set.add_feature(fwd_feature,
color=primer_color,
name=region)
primer_feature_set.add_feature(rev_feature,
color=primer_color,
name=region)
primer_track = GenomeDiagram.Track(name="Annotated Features", height=1)
primer_track.add_set(primer_feature_set)
gd_diagram.add_track(primer_track, 2)
gd_diagram.add_track(gc_track, 1)
rows = max(2, int(round(len(self.primary_ref) / 10000.0)))
gd_diagram.draw(
format="linear",
pagesize=(300 * rows, 200 * rows),
fragments=rows,
start=0,
end=len(self.primary_ref),
)
pdf_filepath = self.outpath / f"{self.prefix}.plot.pdf"
svg_filepath = self.outpath / f"{self.prefix}.plot.svg"
logger.info(f"Writing {pdf_filepath}")
logger.info(f"Writing {svg_filepath}")
gd_diagram.write(str(pdf_filepath), "PDF", dpi=300)
gd_diagram.write(str(svg_filepath), "SVG", dpi=300)
# from a specified genbank file input
from reportlab.lib import colors
from reportlab.lib.units import cm
from Bio.Graphics import GenomeDiagram
from Bio.SeqFeature import SeqFeature, FeatureLocation
from Bio import SeqIO
record1 = input("What is your genbank filename? Ex: ""sequence.gb"" \n")
record = SeqIO.read(record1, "genbank")
pgene_start = int(input("What is the start location of your putative gene?\n"))
pgene_end = int(input("What is the end location of your putative gene?\n"))
pgene_ori = int(input("Is putative gene forward (""1"") or reverse (""-1"")?\n"))
pgene = str(input("What would you like to name your putative gene?\n"))
# create an empty diagram, then add an empty track & empty feature set
gd_diagram = GenomeDiagram.Diagram("S. cerevisiae Chromosome IX")
gd_track_for_features = gd_diagram.new_track(1, name="Annotated Features")
gd_feature_set = gd_track_for_features.new_set()
#Take each gene SeqFeature object in our SeqRecord, and use it to
# generate a feature on the diagram.
for feature in record.features:
if feature.type != "gene":
#Exclude this feature
continue
if len(gd_feature_set) % 2 == 0:
color = colors.lightblue
else:
color = colors.blue
gd_feature_set.add_feature(feature, sigil="ARROW", color=color, arrowshaft_height=1.0,
label=True, label_size=8, label_angle=30)
seqCount = 0
longest = 0
for record in SeqIO.parse(handle, "fasta"):
seqId = record.id
seqLen = len(record)
seqLenMap[seqId] = seqLen
longest = max(longest, seqLen)
seqCount += 1
handle.close()
colorMap = [color_code(i, len(motifMap))for i in range(len(motifMap))]
seqColor = colors.grey.clone(alpha=0.2)
posBED_FH = open(os.path.join(outDir, "sequences_cluster_match_position.bed"), "r")
gdd = GenomeDiagram.Diagram()
prevSeqId = ""
trackId = 0
for line in posBED_FH:
f = line.rstrip('\n').split('\t')
seqId = f[0]
seqLen = int(seqLenMap[seqId])
padLen = max(0, int((longest - seqLen) / 2))
start = int(f[1]) + padLen
end = int(f[2]) + padLen
cluster_id = int(f[3])
if prevSeqId != seqId:
gd_track = gdd.new_track(2 * trackId,
greytrack=True,
start=0,
# to parse the data
from Bio import SeqIO
from Bio.Graphics import GenomeDiagram
#to present the data
from reportlab.lib import colors
from reportlab.lib.units import cm
color_set = [
colors.green, colors.orange, colors.red, colors.purple, colors.cyan
]
record = SeqIO.read("Genome.gb", "genbank")
gd_diagram = GenomeDiagram.Diagram("Tomato Curly Stunt Virus")
gd_track_for_features = gd_diagram.new_track(1, name="Annotated Features")
gd_feature_set = gd_track_for_features.new_set()
for feature in record.features:
if feature.type != "gene":
# dont consider the feature since not a gene.
continue
color = color_set[(len(gd_feature_set))]
# the parameters for representation of each feature.
gd_feature_set.add_feature(feature,
sigil="ARROW",
arrowshaft_height=0.5,
color=color,
label=True,
label_size=25,
def hyperdraw(genome, dataset, label, intensity=False):
for genoslice in genome:
step = 50000
print(len(genoslice), ' in ', step, ' is ', len(genoslice) / step)
gd_diagram = GenomeDiagram.Diagram(genoslice.id)
gd_track_for_features = gd_diagram.new_track(
1,
name="Annotated Features",
scale_ticks=1,
scale_largetick_interval=1000,
scale_smalltick_interval=100,
scale_smallticks=0.05,
scale_largeticks=0.2,
scale_smalltick_labels=0)
gd_feature_set = gd_track_for_features.new_set()
for feature in genoslice.features:
#if feature.type == "operon":
# gd_feature_set.add_feature(feature, sigil="BOX", ##pointy boxes
# color=colors.grey, label=False)
if feature.type == "rRNA" or feature.type == "tRNA":
gd_feature_set.add_feature(
feature,
sigil="OCTO", ##pointy boxes
color=colors.grey)
elif feature.type == "CDS":
# feature.qualifiers['product'][0].lower().find('hypo')>-1:
# feature.qualifiers['product'][0].lower().find('transposase')>-1:
locus = feature.qualifiers['locus_tag'][0]
red, green, blue = (1, 1, 1)
border = colors.gainsboro
ah = 0
if locus in dataset:
if intensity:
intense = float(dataset[locus][1])
red, green = (1 - intense / 10, 1 - intense / 10)
else:
ah = dataset[locus][3]
if ah < 0:
ah = 0
elif ah > 1:
ah = 1
change = float(dataset[locus][1])
sig = float(dataset[locus][2])
if change > 0:
red, blue = (1 - change / 10, 1 - change / 10)
else:
green, blue = (1 - abs(change / 10),
1 - abs(change / 10))
if sig < 0.05:
border = colors.black
else:
border = colors.gray
labellable = True
else:
labellable = False
if len(feature) < 200:
labellable = False
color = colors.Color(red, green, blue)
#shortened=feature.qualifiers['locus_tag'][0]+' '+feature.qualifiers['product'][0]
#if len(shortened)>20:
# shortened=shortened[0:20]+'...'
shortened = feature.qualifiers['product'][0]
for badword in 'hypothetical putative probable family domain unknown possible partial'.split(
):
if feature.qualifiers['product'][0].lower().find(
badword) > -1:
shortened = feature.qualifiers['locus_tag'][0]
allocated = math.floor(len(feature) / 1000 * 30)
if len(shortened) > allocated:
shortened = shortened[0:allocated] + '...'
gd_feature_set.add_feature(
feature,
sigil="ARROW",
arrowshaft_height=ah, ##pointy boxes=1
color=color,
label=labellable,
height=0.8, #not the actually setting...
name=shortened,
label_position="start",
border=border,
#label_strand=1,
label_size=8,
label_angle=0)
i = 0
for x in range(step + 1, len(genoslice) - 100, step):
print(x - step, x + 100)
gs = genoslice[x - step:x + 100]
gd_diagram.draw(format="linear",
pagesize='A4',
orientation='landscape',
fragments=10,
start=x - step,
end=x + 100)
print(gs.features[1].qualifiers['locus_tag'][0],
gs.features[-1].qualifiers['locus_tag'][0])
op = 'C:\\Users\\Cass\\Desktop\\test2\\' + label + gs.features[
1].qualifiers['locus_tag'][0] + '-' + gs.features[
-1].qualifiers['locus_tag'][0]
gd_diagram.write(op + ".pdf", "PDF")
gd_diagram.write(op + ".eps", "EPS")
gd_diagram.write(op + ".svg", "SVG")
gd_diagram.write(op + ".png", "PNG")
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")
def run_module(self):
if self.id_list and os.access(self.id_list[0], os.R_OK):
print("Detected supplied circRNA ID file.")
with open(self.id_list[0]) as f:
lines = f.read().splitlines()
self.id_list = lines
# let's first check if the temporary directory exists
if not (os.access(self.temp_dir, os.W_OK)):
print("Temporary directory %s not writable." % self.temp_dir)
# exit with -1 error if we can't use it
exit(-1)
# let's first check if the temporary directory exists
if not (os.access(self.output_dir, os.W_OK)):
print("Output directory %s not writable." % self.output_dir)
# exit with -1 error if we can't use it
exit(-1)
circ_rna_number = 0
# define temporary files
exon_storage_tmp = self.temp_dir + "circtools_flanking_exons.tmp"
blast_storage_tmp = self.temp_dir + "circtools_blast_results.tmp"
blast_xml_tmp = self.temp_dir + "circtools_blast_results.xml"
output_html_file = self.output_dir + self.experiment_title.replace(
" ", "_") + ".html"
# erase old contents
open(exon_storage_tmp, 'w').close()
# define cache dicts
exon_cache = {}
flanking_exon_cache = {}
primer_to_circ_cache = {}
if self.input_circRNA:
from Bio import SeqIO
with open(exon_storage_tmp, 'a') as data_store:
for record in SeqIO.parse(self.input_circRNA, "fasta"):
# from the FASTA file we cannot tell the coordinates of the circRNA
name = str(record.id) + "_0_0_" + str(len(
record.seq)) + "_0"
data_store.write("\t".join(
[name, str(record.seq), "", "\n"]))
exon_cache[name] = {1: str(record.seq), 2: ""}
else:
exons = self.read_annotation_file(self.gtf_file, entity="exon")
with open(self.dcc_file) as fp:
for line in fp:
# make sure we remove the header
if line.startswith('Chr\t'):
continue
line = line.rstrip()
current_line = line.split('\t')
if current_line[3] == "not_annotated":
continue
if self.gene_list and not self.id_list and current_line[
3] not in self.gene_list:
continue
sep = "_"
name = sep.join([
current_line[3], current_line[0], current_line[1],
current_line[2], current_line[5]
])
if self.id_list and not self.gene_list and name not in self.id_list:
continue
flanking_exon_cache[name] = {}
sep = "\t"
bed_string = sep.join([
current_line[0], current_line[1], current_line[2],
current_line[3],
str(0), current_line[5]
])
virtual_bed_file = pybedtools.BedTool(bed_string,
from_string=True)
result = exons.intersect(virtual_bed_file, s=True)
fasta_bed_line_start = ""
fasta_bed_line_stop = ""
start = 0
stop = 0
for result_line in str(result).splitlines():
bed_feature = result_line.split('\t')
# this is a single-exon circRNA
if bed_feature[1] == current_line[1] and bed_feature[
2] == current_line[2]:
fasta_bed_line_start += result_line + "\n"
start = 1
stop = 1
if bed_feature[1] == current_line[1] and start == 0:
fasta_bed_line_start += result_line + "\n"
start = 1
if bed_feature[2] == current_line[2] and stop == 0:
fasta_bed_line_stop += result_line + "\n"
stop = 1
# these exons are kept for correctly drawing the circRNAs later
# not used for primer design
if bed_feature[1] > current_line[1] and bed_feature[
2] < current_line[2]:
flanking_exon_cache[name][bed_feature[1] + "_" +
bed_feature[2]] = 1
virtual_bed_file_start = pybedtools.BedTool(
fasta_bed_line_start, from_string=True)
virtual_bed_file_stop = pybedtools.BedTool(
fasta_bed_line_stop, from_string=True)
virtual_bed_file_start = virtual_bed_file_start.sequence(
fi=self.fasta_file)
virtual_bed_file_stop = virtual_bed_file_stop.sequence(
fi=self.fasta_file)
if stop == 0 or start == 0:
print(
"Could not identify the exact exon-border of the circRNA."
)
print(
"Will continue with non-annotated, manually extracted sequence."
)
# we have to manually reset the start position
fasta_bed_line = "\t".join([
current_line[0], current_line[1], current_line[2],
current_line[5]
])
virtual_bed_file_start = pybedtools.BedTool(
fasta_bed_line, from_string=True)
virtual_bed_file_start = virtual_bed_file_start.sequence(
fi=self.fasta_file)
virtual_bed_file_stop = ""
exon1 = ""
exon2 = ""
if virtual_bed_file_start:
exon1 = open(
virtual_bed_file_start.seqfn).read().split(
"\n", 1)[1].rstrip()
if virtual_bed_file_stop:
exon2 = open(virtual_bed_file_stop.seqfn).read().split(
"\n", 1)[1].rstrip()
circ_rna_number += 1
print("extracting flanking exons for circRNA #",
circ_rna_number,
name,
end="\n",
flush=True)
if exon2 and not exon1:
exon1 = exon2
exon2 = ""
exon_cache[name] = {1: exon1, 2: exon2}
with open(exon_storage_tmp, 'a') as data_store:
data_store.write("\t".join([name, exon1, exon2, "\n"]))
if not exon_cache:
print(
"Could not find any circRNAs matching your criteria, exiting.")
exit(-1)
# need to define path top R wrapper
primer_script = 'circtools_primex_wrapper.R'
# ------------------------------------ run script and check output -----------------------
script_result = os.popen(primer_script + " " + exon_storage_tmp + " " +
str(self.product_range[0]) + "," +
str(self.product_range[1]) + " " +
self.junction).read()
# this is the first time we look through the input file
# we collect the primer sequences and unify everything in one blast query
blast_object_cache = {}
blast_result_cache = {}
blast_input_file = ""
if circ_rna_number < 50:
for line in script_result.splitlines():
entry = line.split('\t')
circular_rna_id = entry[0].split('_')
if entry[1] == "NA":
continue
# only blast 1
elif entry[2] in blast_object_cache and not entry[
1] in blast_object_cache:
blast_input_file += "\n>" + entry[1] + "\n" + entry[1]
blast_object_cache[entry[1]] = 1
primer_to_circ_cache[entry[1]] = circular_rna_id[0]
# only blast 2
elif entry[1] in blast_object_cache and not entry[
2] in blast_object_cache:
blast_input_file += "\n>" + entry[2] + "\n" + entry[2]
blast_object_cache[entry[2]] = 1
primer_to_circ_cache[entry[2]] = circular_rna_id[0]
# seen both already, skip
elif entry[1] in blast_object_cache and entry[
2] in blast_object_cache:
continue
# nothing seen yet, blast both
else:
blast_input_file += "\n>" + entry[1] + "\n" + entry[
1] + "\n>" + entry[2] + "\n" + entry[2]
blast_object_cache[entry[1]] = 1
blast_object_cache[entry[2]] = 1
primer_to_circ_cache[entry[1]] = circular_rna_id[0]
primer_to_circ_cache[entry[2]] = circular_rna_id[0]
else:
print("Too many circRNAs selected, skipping BLAST step.")
if self.no_blast:
print("User disabled BLAST search, skipping.")
run_blast = 0
# check if we have to blast
if not self.no_blast and blast_input_file:
try:
print("Sending " + str(len(blast_object_cache)) +
" primers to BLAST")
print("This may take a few minutes, please be patient.")
result_handle = self.call_blast(blast_input_file,
self.organism)
run_blast = 1
except Exception as exc:
print(exc)
exit(-1)
with open(blast_xml_tmp, "w") as out_handle:
out_handle.write(result_handle.read())
result_handle.close()
result_handle = open(blast_xml_tmp)
blast_records = NCBIXML.parse(result_handle)
for blast_record in blast_records:
if blast_record.query not in blast_result_cache:
blast_result_cache[blast_record.query] = []
for description in blast_record.descriptions:
# filter out the host gene we're in now
# also filter out all "PREDICTED" stuff
if description.title.find(primer_to_circ_cache[blast_record.query]) == -1 and\
description.title.find("PREDICTED") == -1:
blast_result_cache[blast_record.query].append(
description.title)
# if we encounter NAs nothing has been blasted, we manually set the values now
blast_result_cache["NA"] = ["Not blasted, no primer pair found"]
primex_data_with_blast_results = ""
for line in script_result.splitlines():
entry = line.split('\t')
# split up the identifier for final plotting
line = line.replace("_", "\t")
if run_blast == 1:
left_result = "No hits"
right_result = "No hits"
else:
left_result = "Not blasted, no primer pair found"
right_result = left_result
if entry[1] in blast_result_cache:
left_result = ";".join(blast_result_cache[entry[1]])
if entry[2] in blast_result_cache:
right_result = ";".join(blast_result_cache[entry[2]])
# update line
primex_data_with_blast_results += line + "\t" + left_result + "\t" + right_result + "\n"
with open(blast_storage_tmp, 'w') as data_store:
data_store.write(primex_data_with_blast_results)
# need to define path top R wrapper
primer_script = 'circtools_primex_formatter.R'
# ------------------------------------ run script and check output -----------------------
primex_data_formatted = os.popen(primer_script + " " +
blast_storage_tmp + " " + "\"" +
self.experiment_title + "\"").read()
with open(output_html_file, 'w') as data_store:
data_store.write(primex_data_formatted)
print("Writing results to " + output_html_file)
# here we create the circular graphics for primer visualisation
for line in primex_data_with_blast_results.splitlines():
entry = line.split('\t')
# no primers, no graphics
if entry[6] == "NA":
continue
circular_rna_id = "_".join(
[entry[0], entry[1], entry[2], entry[3], entry[4]])
if circular_rna_id in exon_cache:
circular_rna_id_isoform = circular_rna_id + "_" + entry[5]
circrna_length = int(entry[3]) - int(entry[2])
exon1_length = len(exon_cache[circular_rna_id][1])
exon2_length = len(exon_cache[circular_rna_id][2])
exon2_colour = "#ffac68"
if exon2_length == 0:
exon1_length = int(
len(exon_cache[circular_rna_id][1]) / 2) + 1
exon2_length = int(len(exon_cache[circular_rna_id][1]) / 2)
exon2_colour = "#ff6877"
forward_primer_start = int(
entry[8].split(',')[0]) + circrna_length - exon2_length
forward_primer_length = int(entry[8].split(',')[1])
reverse_primer_start = int(
entry[9].split(',')[0]) - exon2_length
reverse_primer_length = int(entry[9].split(',')[1])
product_size = entry[14]
gdd = GenomeDiagram.Diagram('circRNA primer diagram')
gdt_features = gdd.new_track(
1,
greytrack=True,
name="",
)
gds_features = gdt_features.new_set()
feature = SeqFeature(FeatureLocation(0, exon1_length),
strand=+1)
gds_features.add_feature(feature,
name="Exon 1",
label=False,
color="#ff6877",
label_size=22)
feature = SeqFeature(FeatureLocation(
circrna_length - exon2_length, circrna_length),
strand=+1)
gds_features.add_feature(feature,
name="Exon 2",
label=False,
color=exon2_colour,
label_size=22)
feature = SeqFeature(FeatureLocation(forward_primer_start,
circrna_length),
strand=-1)
gds_features.add_feature(feature,
name="Product",
label=False,
color="#6881ff")
feature = SeqFeature(FeatureLocation(0, reverse_primer_start),
strand=-1)
gds_features.add_feature(feature,
name="Product: " + product_size +
"bp",
label=False,
color="#6881ff",
label_size=22,
label_position="middle")
if self.junction == "f":
feature = SeqFeature(FeatureLocation(
reverse_primer_start - reverse_primer_length,
reverse_primer_start),
strand=-1)
gds_features.add_feature(feature,
name="Reverse",
label=False,
sigil="BIGARROW",
color="#75ff68",
arrowshaft_height=0.3,
arrowhead_length=0.1,
label_size=22)
# the primer spans the BSJ, therefore we have to draw it in two pieces:
# piece 1: primer start to circRNA end
# piece 2: remaining primer portion beginning from 0
# piece 1:
feature = SeqFeature(
FeatureLocation(forward_primer_start, circrna_length))
gds_features.add_feature(feature,
name="Forward",
label=False,
sigil="BIGARROW",
color="#75ff68",
arrowshaft_height=0.3,
arrowhead_length=0.1,
label_size=22)
# piece 2:
feature = SeqFeature(
FeatureLocation(
0, forward_primer_length -
(circrna_length - forward_primer_start)))
gds_features.add_feature(feature,
name="Forward",
label=False,
sigil="BIGARROW",
color="#75ff68",
arrowshaft_height=0.3,
arrowhead_length=0.1,
label_size=22)
elif self.junction == "r":
# the primer spans the BSJ, therefore we have to draw it in two pieces:
# piece 1: primer start of circRNA to circRNA end
# piece 2: remaining primer portion beginning from 0
# piece 1:
feature = SeqFeature(FeatureLocation(
circrna_length - reverse_primer_start, circrna_length),
strand=-1)
gds_features.add_feature(feature,
name="Reverse",
label=False,
sigil="BIGARROW",
color="#75ff68",
arrowshaft_height=0.3,
arrowhead_length=0.1,
label_size=22)
# piece 2:
feature = SeqFeature(FeatureLocation(
0, reverse_primer_start),
strand=-1)
gds_features.add_feature(feature,
name="Reverse",
label=False,
sigil="BIGARROW",
color="#75ff68",
arrowshaft_height=0.3,
arrowhead_length=0.1,
label_size=22)
feature = SeqFeature(
FeatureLocation(
forward_primer_start,
forward_primer_start + forward_primer_length))
gds_features.add_feature(feature,
name="Forward",
label=False,
sigil="BIGARROW",
color="#75ff68",
arrowshaft_height=0.3,
arrowhead_length=0.1,
label_size=22)
else:
feature = SeqFeature(FeatureLocation(
reverse_primer_start - reverse_primer_length,
reverse_primer_start),
strand=-1)
gds_features.add_feature(feature,
name="Reverse",
label=False,
sigil="BIGARROW",
color="#75ff68",
arrowshaft_height=0.3,
arrowhead_length=0.1,
label_size=22)
feature = SeqFeature(
FeatureLocation(
forward_primer_start,
forward_primer_start + forward_primer_length))
gds_features.add_feature(feature,
name="Forward",
label=False,
sigil="BIGARROW",
color="#75ff68",
arrowshaft_height=0.3,
arrowhead_length=0.1,
label_size=22)
feature = SeqFeature(FeatureLocation(0, 1))
gds_features.add_feature(feature,
name="BSJ",
label=True,
color="white",
label_size=22)
if circular_rna_id in flanking_exon_cache:
for exon in flanking_exon_cache[circular_rna_id]:
exon_start, exon_stop = exon.split('_')
exon_start = int(exon_start) - int(entry[2])
exon_stop = int(exon_stop) - int(entry[2])
feature = SeqFeature(FeatureLocation(
exon_start, exon_stop),
strand=+1)
gds_features.add_feature(feature,
name="Exon",
label=False,
color="grey",
label_size=22)
gdd.draw(format='circular',
pagesize=(600, 600),
circle_core=0.6,
track_size=0.3,
tracklines=0,
x=0.00,
y=0.00,
start=0,
end=circrna_length - 1)
gdd.write(
self.output_dir + "/" + circular_rna_id_isoform + ".svg",
"svg")
def write_schemadelica_plot(self, path='./'):
logger.info('Writing plot')
gd_diagram = GenomeDiagram.Diagram("Primer Scheme", track_size=1)
scale_track = GenomeDiagram.Track(name='scale',
scale=True,
scale_fontsize=10,
scale_largetick_interval=1000,
height=0.1)
gd_diagram.add_track(scale_track, 2)
primer_feature_set_1 = GenomeDiagram.FeatureSet()
primer_feature_set_2 = GenomeDiagram.FeatureSet()
for r in self.regions:
cols1 = [
self.primary_reference.id, r.top_pair.left.start,
r.top_pair.left.end, r.top_pair.left.name, r.pool
]
cols2 = [
self.primary_reference.id, r.top_pair.right.end,
r.top_pair.right.start, r.top_pair.right.name, r.pool
]
region = str(r.region_num)
fwd_feature = SeqFeature(
FeatureLocation(int(cols1[1]), int(cols1[2]), strand=0))
rev_feature = SeqFeature(
FeatureLocation(int(cols2[1]), int(cols2[2]), strand=0))
region_feature = SeqFeature(
FeatureLocation(int(cols1[1]), int(cols2[2]), strand=0))
if int(region) % 2 == 0:
primer_feature_set_1.add_feature(region_feature,
color=colors.palevioletred,
name=region,
label=True,
label_size=10,
label_position="middle",
label_angle=0)
primer_feature_set_1.add_feature(fwd_feature,
color=colors.red,
name=region,
label=False)
primer_feature_set_1.add_feature(rev_feature,
color=colors.red,
name=region,
label=False)
else:
primer_feature_set_2.add_feature(region_feature,
color=colors.palevioletred,
name=region,
label=True,
label_size=10,
label_position="middle",
label_angle=0)
primer_feature_set_2.add_feature(fwd_feature,
color=colors.red,
name=region,
label=False)
primer_feature_set_2.add_feature(rev_feature,
color=colors.red,
name=region,
label=False)
primer_track = GenomeDiagram.Track(name="Annotated Features",
height=0.1)
primer_track.add_set(primer_feature_set_1)
gd_diagram.add_track(primer_track, 4)
primer_track = GenomeDiagram.Track(name="Annotated Features",
height=0.1)
primer_track.add_set(primer_feature_set_2)
gd_diagram.add_track(primer_track, 6)
rows = max(2, int(round(len(self.primary_reference) / 10000.0)))
gd_diagram.draw(format='linear',
pagesize=(300 * rows, 200 * rows),
fragments=rows,
start=0,
end=len(self.primary_reference))
png_filepath = os.path.join(path, '{}.png'.format(self.prefix))
pdf_filepath = os.path.join(path, '{}.pdf'.format(self.prefix))
svg_filepath = os.path.join(path, '{}.svg'.format(self.prefix))
gd_diagram.write(png_filepath, 'PNG', dpi=300)
gd_diagram.write(pdf_filepath, 'PDF', dpi=300)
gd_diagram.write(svg_filepath, 'SVG', dpi=300)
def export_plasmidmap(gbfile, filename=None):
""" Export Linear and Circular Plasmid Map for the imported GenBank file.
:param gbfile: a genbank file in .gb format or the path the file if not in the
same folder.
:type gbfile: str
:param filename: the filenames/path to the filenames for the linear and
circular plasmids in tuple
:type filename: tuple, optional
:return: the version from the genbank file
:rtype: str
"""
record = SeqIO.read(gbfile, "genbank")
gd_diagram = GenomeDiagram.Diagram(record.id)
gd_track_for_features = gd_diagram.new_track(1,
name="Annotated Features")
gd_feature_set = gd_track_for_features.new_set()
for feature in record.features:
if feature.type == "primer" or (feature.type == "misc_feature"):
continue
# if (feature.type != "CDS"):
# # Exclude this feature
# continue
if len(gd_feature_set) % 2 == 0:
color = colors.lightblue
else:
color = colors.blue
gd_feature_set.add_feature(
feature,
sigil="ARROW",
color=color,
label_size=12,
label_angle=0,
label=True,
)
# Draw Linear map from genbank
gd_diagram.draw(
format="linear",
orientation="landscape",
pagesize="A4",
fragments=4,
start=0,
end=len(record),
)
if filename is None:
linfile = "plasmid_linear.png"
circfile = "plasmid_circular.png"
else:
linfile = filename[0]
circfile = filename[1]
gd_diagram.write(linfile, "PNG")
# Draw circular map from genbank
gd_diagram.draw(
format="circular",
circular=True,
pagesize=(25 * cm, 20 * cm), # pagesize=(35 * cm, 30 * cm),
start=0,
end=len(record),
circle_core=0.5,
)
# gd_diagram.write("plasmid_circular.pdf", "PDF")
gd_diagram.write(circfile, "PNG")
return record.id
scaffold2orf2coord[ scaffold ][orf] = (int(start),int(end),int(strand))
print(scaffold2orf2coord[ scaffold ][orf])
file.close()
from reportlab.lib import colors
from reportlab.lib.units import cm
from Bio.Graphics import GenomeDiagram
from Bio import SeqIO
for scaffold,orf2coord in scaffold2orf2coord.items() :
print(scaffold)
scaffold_filename = scaffold+'_genomicDiagram.pdf'
gd_diagram = GenomeDiagram.Diagram("a new type of nitrogenase")
gd_track_for_features = gd_diagram.new_track(1, name="Annotated Features")
gd_feature_set = gd_track_for_features.new_set()
# print('\t'+str(scaffold2max[scaffold]))
# print('\t'+str(scaffold2min[scaffold]))
for orf,coord in orf2coord.items() :
start = coord[0]
end = coord[1]
strand = coord[2]
feature = SeqFeature( FeatureLocation( int( int(start) ) , int( int(end) ) ) , strand=int(strand), type = 'CDS' )
if orf in orf2color :
color = orf2color[orf]
else:
from Bio.Graphics import GenomeDiagram
from reportlab.lib import colors
from reportlab.lib.units import cm
from Bio import SeqIO, SeqFeature
from Bio.SeqFeature import SeqFeature, FeatureLocation
#this code is from https://biopython-tutorial.readthedocs.io/en/latest/notebooks/17%20-%20Graphics%20including%20GenomeDiagram.html
#I used it, and that web page, to learn how to use GenomeDiagram from Biopython
#define parameters of drawing
gdd = GenomeDiagram.Diagram('Test Diagram')
gdt_features = gdd.new_track(1, greytrack=False)
gds_features = gdt_features.new_set()
#Add three features to show the strand options,
feature = SeqFeature(FeatureLocation(25, 125), strand=+1)
gds_features.add_feature(feature, name="Forward", label=True, sigil="ARROW")
feature = SeqFeature(FeatureLocation(150, 250), strand=None)
gds_features.add_feature(feature, name="Strandless", label=True, sigil="ARROW")
feature = SeqFeature(FeatureLocation(275, 375), strand=-1)
gds_features.add_feature(feature, name="Reverse", label=True, sigil="ARROW")
#draw and save diagram
gdd.draw(format='linear',
pagesize=(15 * cm, 4 * cm),
fragments=1,
start=0,
end=400)
gdd.write("GD_labels_default.png", "png")
# BLAST
SeqIO.convert(A, "genbank", A + ".fasta", "fasta")
SeqIO.convert(B, "genbank", B + ".fasta", "fasta")
comando_blastn = NcbiblastnCommandline( \
query=A+".fasta", subject=B+".fasta", \
outfmt='6 qstart qend sstart send pident',\
out="blast_"+A+"_"+B+".txt")
stdout, stderr = comando_blastn()
blast = open("blast_" + A + "_" + B + ".txt")
# Iniciando a figura
name = A + "_" + B
gd = GenomeDiagram.Diagram(name)
gA = gd.new_track(1,name="A",height=0.5, \
start=0,end=len(A1))
gA1 = gA.new_set()
gB = gd.new_track(3,name="B",height=0.5, \
start=0,end=len(B1))
gB1 = gB.new_set()
# Cores CDSs - intercalado
c1 = "#79B134"
c2 = "#8DE91D"
# Colore um quadrado para cada CDS do arquivo A
cont = 1
for i in A1.features:
p5 = []
p6 = []
for p,pp in zip(p1,p2):
for ppp,pppp in zip(p3,p4):
if ppp < p < pppp:
p5.append(p)
p6.append(pppp)
elif p < ppp < pp:
p5.append(ppp)
p6.append(pp)
for p,pp, i in zip(p5,p6,range(1,len(p5)+1)):
record.features.append(SeqFeature(location = FeatureLocation(p, pp, strand = None), type = 'overlap', id = 'o{}'.format(i)))
gd_diagram = GenomeDiagram.Diagram("Promoter region with GRE and TF binding")
gd_track_for_features = gd_diagram.new_track(1, name="GREs and TFBS")
gd_feature_set = gd_track_for_features.new_set()
for feature in record.features:
if feature.type == "GRE":
color = colors.green
gd_feature_set.add_feature(feature, color = color, label = feature.id, label_size=12, label_angle=20, sigil = "OCTO")
if feature.type == "TF":
color = colors.red
gd_feature_set.add_feature(feature, color = color, label = feature.id, label_size=12, label_angle=20, sigil = "OCTO")
# else:
# gd_feature_set.add_feature(feature, color = colors.yellow, label = False)
gd_diagram.draw(format="linear", orientation="landscape", pagesize=(300,80),
fragments=1, start=0, end=len(record))
