def gaps(args):
"""
%prog gaps A_vs_B.blast
Find distribution of gap sizes betwen adjacent HSPs.
"""
p = OptionParser(gaps.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
blastfile, = args
blast = BlastSlow(blastfile)
logging.debug("A total of {} records imported".format(len(blast)))
query_gaps = list(collect_gaps(blast))
subject_gaps = list(collect_gaps(blast, use_subject=True))
logging.debug("Query gaps: {} Subject gaps: {}"\
.format(len(query_gaps), len(subject_gaps)))
from jcvi.graphics.base import savefig
import seaborn as sns
sns.distplot(query_gaps)
savefig("query_gaps.pdf")
def coverage(args):
"""
%prog coverage *.coverage
Plot coverage along chromosome. The coverage file can be generated with:
$ samtools depth a.bam > a.coverage
The plot is a simple line plot using matplotlib.
"""
from jcvi.graphics.base import savefig
p = OptionParser(coverage.__doc__)
opts, args, iopts = p.set_image_options(args, format="png")
if len(args) != 1:
sys.exit(not p.print_help())
covfile, = args
df = pd.read_csv(covfile, sep='\t', names=["Ref", "Position", "Depth"])
xlabel, ylabel = "Position", "Depth"
df.plot(xlabel, ylabel, color='g')
image_name = covfile + "." + iopts.format
savefig(image_name)
def draw(self, title="Ks distribution"):
from jcvi.graphics.base import tex_formatter, \
tex_1digit_formatter, tex_2digit_formatter, savefig
ax = self.ax
ks_max = self.ks_max
lines = self.lines
labels = self.labels
legendp = self.legendp
leg = ax.legend(lines, labels, legendp,
shadow=True, fancybox=True, prop={"size": 10})
leg.get_frame().set_alpha(.5)
ax.set_xlim((0, ks_max - self.interval))
ax.set_title(title, fontweight="bold")
ax.set_xlabel('Synonymous substitutions per site (Ks)')
ax.set_ylabel('Percentage of gene pairs')
tf = tex_2digit_formatter if self.interval < .1 else \
tex_1digit_formatter
ax.xaxis.set_major_formatter(tf)
ax.yaxis.set_major_formatter(tex_formatter)
image_name = "Ks_plot.pdf"
savefig(image_name, dpi=300)
def birch(args):
"""
%prog birch seqids layout
Plot birch macro-synteny, with an embedded phylogenetic tree to the right.
"""
p = OptionParser(birch.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="8x6")
if len(args) != 2:
sys.exit(not p.print_help())
seqids, layout = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
K = Karyotype(fig, root, seqids, layout)
L = K.layout
xs = .79
dt = dict(rectangle=False, circle=False)
# Embed a phylogenetic tree to the right
coords = {}
coords["Amborella"] = (xs, L[0].y)
coords["Vitis"] = (xs, L[1].y)
coords["Prunus"] = (xs, L[2].y)
coords["Betula"] = (xs, L[3].y)
coords["Populus"] = (xs, L[4].y)
coords["Arabidopsis"] = (xs, L[5].y)
coords["fabids"] = join_nodes(root, coords, "Prunus", "Betula", xs, **dt)
coords["malvids"] = join_nodes(root, coords, \
"Populus", "Arabidopsis", xs, **dt)
coords["rosids"] = join_nodes(root, coords, "fabids", "malvids", xs, **dt)
coords["eudicots"] = join_nodes(root, coords, "rosids", "Vitis", xs, **dt)
coords["angiosperm"] = join_nodes(root, coords, \
"eudicots", "Amborella", xs, **dt)
# Show branch length
branch_length(root, coords["Amborella"], coords["angiosperm"], ">160.0")
branch_length(root, coords["eudicots"], coords["angiosperm"],
">78.2", va="top")
branch_length(root, coords["Vitis"], coords["eudicots"], "138.5")
branch_length(root, coords["rosids"], coords["eudicots"],
"19.8", va="top")
branch_length(root, coords["Prunus"], coords["fabids"],
"104.2", ha="right", va="top")
branch_length(root, coords["Arabidopsis"], coords["malvids"],
"110.2", va="top")
branch_length(root, coords["fabids"], coords["rosids"],
"19.8", ha="right", va="top")
branch_length(root, coords["malvids"], coords["rosids"],
"8.5", va="top")
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "birch"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def resample(args):
"""
%prog resample yellow-catfish-resample.txt medicago-resample.txt
Plot ALLMAPS performance across resampled real data.
"""
p = OptionParser(resample.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="8x4", dpi=300)
if len(args) != 2:
sys.exit(not p.print_help())
dataA, dataB = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
A = fig.add_axes([.1, .18, .32, .64])
B = fig.add_axes([.6, .18, .32, .64])
dataA = import_data(dataA)
dataB = import_data(dataB)
xlabel = "Fraction of markers"
ylabels = ("Anchor rate", "Runtime (m)")
legend = ("anchor rate", "runtime")
subplot_twinx(A, dataA, xlabel, ylabels,
title="Yellow catfish", legend=legend)
subplot_twinx(B, dataB, xlabel, ylabels,
title="Medicago", legend=legend)
labels = ((.04, .92, "A"), (.54, .92, "B"))
panel_labels(root, labels)
normalize_axes(root)
image_name = "resample." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def pomegranate(args):
"""
%prog cotton seqids karyotype.layout mcscan.out all.bed synteny.layout
Build a figure that calls graphics.karyotype to illustrate the high ploidy
of WGD history of pineapple genome. The script calls both graphics.karyotype
and graphic.synteny.
"""
p = OptionParser(pomegranate.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="9x7")
if len(args) != 5:
sys.exit(not p.print_help())
seqidsfile, klayout, datafile, bedfile, slayout = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Karyotype(fig, root, seqidsfile, klayout)
Synteny(fig, root, datafile, bedfile, slayout)
# legend showing the orientation of the genes
draw_gene_legend(root, .42, .52, .48)
labels = ((.04, .96, 'A'), (.04, .52, 'B'))
panel_labels(root, labels)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "pomegranate-karyotype"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def epoch(args):
"""
%prog epoch
Illustrate the methods used in Maggie's epoch paper, in particular, how to
classifiy S/G/F/FB/FN for the genes.
"""
p = OptionParser(__doc__)
opts, args = p.parse_args()
fig = plt.figure(1, (6, 4))
root = fig.add_axes([0, 0, 1, 1])
# Separators
linestyle = dict(lw=2, color="b", alpha=.2, zorder=2)
root.plot((0, 1), (.5, .5), "--", **linestyle)
for i in (1./3, 2./3):
root.plot((i, i), (.5, 1), "--", **linestyle)
for i in (1./6, 3./6, 5./6):
root.plot((i, i), (0, .5), "--", **linestyle)
# Diagrams
plot_diagram(root, 1./6, 3./4, "S", "syntenic")
plot_diagram(root, 3./6, 3./4, "F", "missing, with both flankers")
plot_diagram(root, 5./6, 3./4, "G", "missing, with one flanker")
plot_diagram(root, 2./6, 1./4, "FB", "has non-coding matches")
plot_diagram(root, 4./6, 1./4, "FN", "syntenic region has gap")
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
figname = fname() + ".pdf"
savefig(figname, dpi=300)
def main():
p = OptionParser(__doc__)
p.add_option("--switch",
help="Rename the seqid with two-column file [default: %default]")
p.add_option("--tree",
help="Display trees on the bottom of the figure [default: %default]")
p.add_option("--extra", help="Extra features in BED format")
p.add_option("--scalebar", default=False, action="store_true",
help="Add scale bar to the plot")
opts, args, iopts = p.set_image_options(figsize="8x7")
if len(args) != 3:
sys.exit(not p.print_help())
datafile, bedfile, layoutfile = args
switch = opts.switch
tree = opts.tree
pf = datafile.rsplit(".", 1)[0]
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Synteny(fig, root, datafile, bedfile, layoutfile,
switch=switch, tree=tree, extra_features=opts.extra,
scalebar=opts.scalebar)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def main():
p = OptionParser(__doc__)
p.add_option("--order",
help="The order to plot the tracks, comma-separated")
opts, args, iopts = p.set_image_options()
if len(args) != 3:
sys.exit(not p.print_help())
chr, sizes, datadir = args
order = opts.order
hlsuffix = opts.hlsuffix
if order:
order = order.split(",")
sizes = Sizes(sizes)
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
canvas = (.12, .35, .8, .35)
chr_size = sizes.get_size(chr)
c = Coverage(fig, root, canvas, chr, (0, chr_size), datadir,
order=order, hlsuffix=hlsuffix)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = chr + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def oropetium(args):
"""
%prog oropetium mcscan.out all.bed layout switch.ids
Build a composite figure that calls graphis.synteny.
"""
p = OptionParser(oropetium.__doc__)
p.add_option("--extra", help="Extra features in BED format")
opts, args, iopts = p.set_image_options(args, figsize="9x6")
if len(args) != 4:
sys.exit(not p.print_help())
datafile, bedfile, slayout, switch = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Synteny(fig, root, datafile, bedfile, slayout,
switch=switch, extra_features=opts.extra)
# legend showing the orientation of the genes
draw_gene_legend(root, .4, .57, .74, text=True, repeat=True)
# On the left panel, make a species tree
fc = 'lightslategrey'
coords = {}
xs, xp = .16, .03
coords["oropetium"] = (xs, .7)
coords["setaria"] = (xs, .6)
coords["sorghum"] = (xs, .5)
coords["rice"] = (xs, .4)
coords["brachypodium"] = (xs, .3)
xs -= xp
coords["Panicoideae"] = join_nodes(root, coords, "setaria", "sorghum", xs)
xs -= xp
coords["BEP"] = join_nodes(root, coords, "rice", "brachypodium", xs)
coords["PACMAD"] = join_nodes(root, coords, "oropetium", "Panicoideae", xs)
xs -= xp
coords["Poaceae"] = join_nodes(root, coords, "BEP", "PACMAD", xs)
# Names of the internal nodes
for tag in ("BEP", "Poaceae"):
nx, ny = coords[tag]
nx, ny = nx - .005, ny - .02
root.text(nx, ny, tag, rotation=90, ha="right", va="top", color=fc)
for tag in ("PACMAD",):
nx, ny = coords[tag]
nx, ny = nx - .005, ny + .02
root.text(nx, ny, tag, rotation=90, ha="right", va="bottom", color=fc)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "oropetium"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def multilineplot(args):
"""
%prog multilineplot fastafile chr1
Combine multiple line plots in one vertical stack
Inputs must be BED-formatted.
--lines: traditional line plots, useful for plotting feature freq
"""
p = OptionParser(multilineplot.__doc__)
p.add_option("--lines",
help="Features to plot in lineplot [default: %default]")
p.add_option("--colors",
help="List of colors matching number of input bed files")
p.add_option("--mode", default="span", choices=("span", "count", "score"),
help="Accumulate feature based on [default: %default]")
p.add_option("--binned", default=False, action="store_true",
help="Specify whether the input is already binned; " +
"if True, input files are considered to be binfiles")
add_window_options(p)
opts, args, iopts = p.set_image_options(args, figsize="8x5")
if len(args) != 2:
sys.exit(not p.print_help())
fastafile, chr = args
window, shift, subtract = check_window_options(opts)
linebeds = []
colors = opts.colors
if opts.lines:
lines = opts.lines.split(",")
assert len(colors) == len(lines), "Number of chosen colors must match" + \
" number of input bed files"
linebeds = get_beds(lines, binned=opts.binned)
linebins = get_binfiles(linebeds, fastafile, shift, mode=opts.mode, binned=opts.binned)
clen = Sizes(fastafile).mapping[chr]
nbins = get_nbins(clen, shift)
plt.rcParams["xtick.major.size"] = 0
plt.rcParams["ytick.major.size"] = 0
plt.rcParams["figure.figsize"] = iopts.w, iopts.h
fig, axarr = plt.subplots(nrows=len(lines))
if len(linebeds) == 1:
axarr = (axarr, )
fig.suptitle(chr, color="darkslategray")
for i, ax in enumerate(axarr):
lineplot(ax, [linebins[i]], nbins, chr, window, shift, \
color="{0}{1}".format(colors[i], 'r'))
plt.subplots_adjust(hspace=0.5)
image_name = chr + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def plot_data(x, y, tour, M):
from jcvi.graphics.base import plt, savefig
plt.plot(x, y, "ro")
for ia, ib in pairwise(tour):
plt.plot((x[ia], x[ib]), (y[ia], y[ib]), "r-")
score = evaluate(tour, M)
plt.title("Score={0:.2f}".format(score))
savefig("demo.pdf")
def venn(args):
"""
%prog venn *.benchmark
Display benchmark results as Venn diagram.
"""
from matplotlib_venn import venn2
p = OptionParser(venn.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="9x9")
if len(args) < 1:
sys.exit(not p.print_help())
bcs = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
pad = .02
ystart = 1
ywidth = 1. / len(bcs)
tags = ("Bowers", "YGOB", "Schnable")
for bc, tag in zip(bcs, tags):
fp = open(bc)
data = []
for row in fp:
prog, pcounts, tcounts, shared = row.split()
pcounts = int(pcounts)
tcounts = int(tcounts)
shared = int(shared)
data.append((prog, pcounts, tcounts, shared))
xstart = 0
xwidth = 1. / len(data)
for prog, pcounts, tcounts, shared in data:
a, b, c = pcounts - shared, tcounts - shared, shared
ax = fig.add_axes([xstart + pad, ystart - ywidth + pad,
xwidth - 2 * pad, ywidth - 2 * pad])
venn2(subsets=(a, b, c), set_labels=(prog, tag), ax=ax)
message = "Sn={0} Pu={1}".\
format(percentage(shared, tcounts, precision=0, mode=-1),
percentage(shared, pcounts, precision=0, mode=-1))
print >> sys.stderr, message
ax.text(.5, .92, latex(message), ha="center", va="center",
transform=ax.transAxes, color='b')
ax.set_axis_off()
xstart += xwidth
ystart -= ywidth
panel_labels(root, ((.04, .96, "A"), (.04, .96 - ywidth, "B"),
(.04, .96 - 2 * ywidth, "C")))
panel_labels(root, ((.5, .98, "A. thaliana duplicates"),
(.5, .98 - ywidth, "14 Yeast genomes"),
(.5, .98 - 2 * ywidth, "4 Grass genomes")))
normalize_axes(root)
savefig("venn.pdf", dpi=opts.dpi)
def scenario(args):
"""
%prog scenario
Illustration of the two-step genome merger process for B. rapa companion paper.
"""
p = OptionParser(__doc__)
opts, args = p.parse_args()
fig = plt.figure(1, (5, 5))
root = fig.add_axes([0, 0, 1, 1])
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
# Layout format: (x, y, label, (chr lengths))
anc = (.5, .9, "Ancestor", (1,))
s1 = (.2, .6, "Genome I", (1,))
s2 = (.5, .6, "Genome II", (1,))
s3 = (.8, .6, "Genome III", (1,))
tetra = (.35, .4, "Tetraploid I / II", (.5, .9))
hexa = (.5, .1, "Hexaploid I / II / III", (.36, .46, .9))
labels = (anc, s1, s2, s3, tetra, hexa)
connections = ((anc, s1), (anc, s2), (anc, s3),\
(s1, tetra), (s2, tetra),
(tetra, hexa), (s3, hexa))
xinterval = .02
yratio = .05
for xx, yy, label, chrl in labels:
#RoundLabel(root, xx, yy, label)
root.text(xx, yy, label, ha="center", va="center")
offset = len(label) * .012
for i, c in enumerate(chrl):
ya = yy + yratio * c
yb = yy - yratio * c
Chromosome(root, xx - offset + i * xinterval, ya, yb, width=.01)
# Comments
comments = ((.15, .33, "II dominant"),
(.25, .03, "III dominant"))
for xx, yy, c in comments:
root.text(xx, yy, c, size=9, ha="center", va="center")
# Branches
tip = .04
for a, b in connections:
xa, ya, la, chra = a
xb, yb, lb, chrb = b
plt.plot((xa, xb), (ya - tip, yb + 2 * tip), 'k-', lw=2, alpha=.5)
figname = fname() + ".pdf"
savefig(figname, dpi=300)
def litchi(args):
"""
%prog litchi mcscan.out all.bed layout switch.ids
Build a composite figure that calls graphis.synteny.
"""
p = OptionParser(litchi.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="9x6")
if len(args) != 4:
sys.exit(not p.print_help())
datafile, bedfile, slayout, switch = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Synteny(fig, root, datafile, bedfile, slayout, switch=switch)
# legend showing the orientation of the genes
draw_gene_legend(root, .4, .7, .82)
# On the left panel, make a species tree
fc = 'lightslategrey'
coords = {}
xs, xp = .16, .03
coords["lychee"] = (xs, .37)
coords["clementine"] = (xs, .5)
coords["cacao"] = (xs, .6)
coords["strawberry"] = (xs, .7)
coords["grape"] = (xs, .8)
xs -= xp
coords["Sapindales"] = join_nodes(root, coords, "clementine", "lychee", xs)
xs -= xp
coords["Rosid-II"] = join_nodes(root, coords, "cacao", "Sapindales", xs)
xs -= xp
coords["Rosid"] = join_nodes(root, coords, "strawberry", "Rosid-II", xs)
xs -= xp
coords["crown"] = join_nodes(root, coords, "grape", "Rosid", xs,
circle=False)
# Names of the internal nodes
for tag in ("Rosid", "Rosid-II", "Sapindales"):
nx, ny = coords[tag]
nx, ny = nx - .01, ny - .02
root.text(nx, ny, tag, rotation=90, ha="right", va="top", color=fc)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "litchi"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def covlen(args):
"""
%prog covlen covfile fastafile
Plot coverage vs length. `covfile` is two-column listing contig id and
depth of coverage.
"""
import numpy as np
import pandas as pd
import seaborn as sns
from jcvi.formats.base import DictFile
p = OptionParser(covlen.__doc__)
p.add_option("--maxsize", default=1000000, type="int", help="Max contig size")
p.add_option("--maxcov", default=100, type="int", help="Max contig size")
p.add_option("--color", default='m', help="Color of the data points")
p.add_option("--kind", default="scatter",
choices=("scatter", "reg", "resid", "kde", "hex"),
help="Kind of plot to draw")
opts, args, iopts = p.set_image_options(args, figsize="8x8")
if len(args) != 2:
sys.exit(not p.print_help())
covfile, fastafile = args
cov = DictFile(covfile, cast=float)
s = Sizes(fastafile)
data = []
maxsize, maxcov = opts.maxsize, opts.maxcov
for ctg, size in s.iter_sizes():
c = cov.get(ctg, 0)
if size > maxsize:
continue
if c > maxcov:
continue
data.append((size, c))
x, y = zip(*data)
x = np.array(x)
y = np.array(y)
logging.debug("X size {0}, Y size {1}".format(x.size, y.size))
df = pd.DataFrame()
xlab, ylab = "Length", "Coverage of depth (X)"
df[xlab] = x
df[ylab] = y
sns.jointplot(xlab, ylab, kind=opts.kind, data=df,
xlim=(0, maxsize), ylim=(0, maxcov),
stat_func=None, edgecolor="w", color=opts.color)
figname = covfile + ".pdf"
savefig(figname, dpi=iopts.dpi, iopts=iopts)
def snpplot(args):
"""
%prog counts.cdt
Illustrate the histogram per SNP site.
"""
p = OptionParser(snpplot.__doc__)
opts, args, iopts = p.set_image_options(args, format="png")
if len(args) != 1:
sys.exit(not p.print_help())
datafile, = args
# Read in CDT file
fp = open(datafile)
next(fp)
next(fp)
data = []
for row in fp:
atoms = row.split()[4:]
nval = len(atoms)
values = [float(x) for x in atoms]
# normalize
values = [x * 1. / sum(values) for x in values]
data.append(values)
pf = datafile.rsplit(".", 1)[0]
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
xmin, xmax = .1, .9
ymin, ymax = .1, .9
yinterval = (ymax - ymin) / len(data)
colors = "rbg" if nval == 3 else ["lightgray"] + list("rbg")
ystart = ymax
for d in data:
xstart = xmin
for dd, c in zip(d, colors):
xend = xstart + (xmax - xmin) * dd
root.plot((xstart, xend), (ystart, ystart), "-", color=c)
xstart = xend
ystart -= yinterval
root.text(.05, .5, "{0} LMD50 SNPs".format(len(data)),
ha="center", va="center", rotation=90, color="lightslategray")
for x, t, c in zip((.3, .5, .7), ("REF", "ALT", "HET"), "rbg"):
root.text(x, .95, t, color=c, ha="center", va="center")
normalize_axes(root)
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def dotplot_main(anchorfile, qbed, sbed, image_name, iopts, vmin=0, vmax=1,
is_self=False, synteny=False, cmap_text=None, cmap="copper", genomenames=None,
sample_number=10000, minfont=5, palette=None, chrlw=.01, title=None):
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1]) # the whole canvas
ax = fig.add_axes([.1, .1, .8, .8]) # the dot plot
dotplot(anchorfile, qbed, sbed, fig, root, ax, vmin=vmin, vmax=vmax,
is_self=is_self, synteny=synteny, cmap_text=cmap_text, cmap=cmap,
genomenames=genomenames, sample_number=sample_number,
minfont=minfont, palette=palette, chrlw=chrlw, title=title)
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def ploidy(args):
"""
%prog cotton seqids karyotype.layout mcscan.out all.bed synteny.layout
Build a figure that calls graphics.karyotype to illustrate the high ploidy
of WGD history of pineapple genome. The script calls both graphics.karyotype
and graphic.synteny.
"""
p = OptionParser(ploidy.__doc__)
p.add_option("--switch", help="Rename the seqid with two-column file")
opts, args, iopts = p.set_image_options(args, figsize="9x7")
if len(args) != 5:
sys.exit(not p.print_help())
seqidsfile, klayout, datafile, bedfile, slayout = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Karyotype(fig, root, seqidsfile, klayout)
Synteny(fig, root, datafile, bedfile, slayout, switch=opts.switch)
# legend showing the orientation of the genes
draw_gene_legend(root, .27, .37, .52)
# annotate the WGD events
fc = 'lightslategrey'
x = .09
radius = .012
TextCircle(root, x, .825, r'$\tau$', radius=radius, fc=fc)
TextCircle(root, x, .8, r'$\sigma$', radius=radius, fc=fc)
TextCircle(root, x, .72, r'$\rho$', radius=radius, fc=fc)
for ypos in (.825, .8, .72):
root.text(.12, ypos, r"$\times2$", color=fc, ha="center", va="center")
root.plot([x, x], [.85, .775], ":", color=fc, lw=2)
root.plot([x, x], [.75, .675], ":", color=fc, lw=2)
labels = ((.04, .96, 'A'), (.04, .54, 'B'))
panel_labels(root, labels)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "pineapple-karyotype"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def amborella(args):
"""
%prog amborella seqids karyotype.layout mcscan.out all.bed synteny.layout
Build a composite figure that calls graphics.karyotype and graphics.synteny.
"""
p = OptionParser(amborella.__doc__)
p.add_option("--tree",
help="Display trees on the bottom of the figure [default: %default]")
p.add_option("--switch",
help="Rename the seqid with two-column file [default: %default]")
opts, args, iopts = p.set_image_options(args, figsize="8x7")
if len(args) != 5:
sys.exit(not p.print_help())
seqidsfile, klayout, datafile, bedfile, slayout = args
switch = opts.switch
tree = opts.tree
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Karyotype(fig, root, seqidsfile, klayout)
Synteny(fig, root, datafile, bedfile, slayout, switch=switch, tree=tree)
# legend showing the orientation of the genes
draw_gene_legend(root, .5, .68, .5)
# annotate the WGD events
fc = 'lightslategrey'
x = .05
radius = .012
TextCircle(root, x, .86, '$\gamma$', radius=radius)
TextCircle(root, x, .95, '$\epsilon$', radius=radius)
root.plot([x, x], [.83, .9], ":", color=fc, lw=2)
pts = plot_cap((x, .95), np.radians(range(-70, 250)), .02)
x, y = zip(*pts)
root.plot(x, y, ":", color=fc, lw=2)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "amborella"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def ploidy(args):
"""
%prog ploidy seqids layout
Build a figure that calls graphics.karyotype to illustrate the high ploidy
of B. napus genome.
"""
p = OptionParser(ploidy.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="8x7")
if len(args) != 2:
sys.exit(not p.print_help())
seqidsfile, klayout = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Karyotype(fig, root, seqidsfile, klayout)
fc = "darkslategrey"
radius = .012
ot = -.05 # use this to adjust vertical position of the left panel
TextCircle(root, .1, .9 + ot, r'$\gamma$', radius=radius, fc=fc)
root.text(.1, .88 + ot, r"$\times3$", ha="center", va="top", color=fc)
TextCircle(root, .08, .79 + ot, r'$\alpha$', radius=radius, fc=fc)
TextCircle(root, .12, .79 + ot, r'$\beta$', radius=radius, fc=fc)
root.text(.1, .77 + ot, r"$\times3\times2\times2$", ha="center", va="top", color=fc)
root.text(.1, .67 + ot, r"Brassica triplication", ha="center",
va="top", color=fc, size=11)
root.text(.1, .65 + ot, r"$\times3\times2\times2\times3$", ha="center", va="top", color=fc)
root.text(.1, .42 + ot, r"Allo-tetraploidy", ha="center",
va="top", color=fc, size=11)
root.text(.1, .4 + ot, r"$\times3\times2\times2\times3\times2$", ha="center", va="top", color=fc)
bb = dict(boxstyle="round,pad=.5", fc="w", ec="0.5", alpha=0.5)
root.text(.5, .2 + ot, r"\noindent\textit{Brassica napus}\\"
"(A$\mathsf{_n}$C$\mathsf{_n}$ genome)", ha="center",
size=16, color="k", bbox=bb)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "napus"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def _draw_trees(trees, nrow=1, ncol=1, rmargin=.3, iopts=None, outdir=".",
shfile=None, **kwargs):
"""
Draw one or multiple trees on one plot.
"""
from jcvi.graphics.tree import draw_tree
if shfile:
SHs = DictFile(shfile, delimiter="\t")
ntrees = len(trees)
n = nrow * ncol
for x in xrange(int(ceil(float(ntrees)/n))):
fig = plt.figure(1, (iopts.w, iopts.h)) if iopts \
else plt.figure(1, (5, 5))
root = fig.add_axes([0, 0, 1, 1])
xiv = 1. / ncol
yiv = 1. / nrow
xstart = list(np.arange(0, 1, xiv)) * nrow
ystart = list(chain(*zip(*[list(np.arange(0, 1, yiv))[::-1]] * ncol)))
for i in xrange(n*x, n*(x+1)):
if i == ntrees:
break
ax = fig.add_axes([xstart[i%n], ystart[i%n], xiv, yiv])
f = trees.keys()[i]
tree = trees[f]
try:
SH = SHs[f]
except:
SH = None
draw_tree(ax, tree, rmargin=rmargin, reroot=False, \
supportcolor="r", SH=SH, **kwargs)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
format = iopts.format if iopts else "pdf"
dpi = iopts.dpi if iopts else 300
if n == 1:
image_name = f.rsplit(".", 1)[0] + "." + format
else:
image_name = "trees{0}.{1}".format(x, format)
image_name = op.join(outdir, image_name)
savefig(image_name, dpi=dpi, iopts=iopts)
plt.clf()
def correlation(args):
"""
%prog correlation postgenomic-s.tsv
Plot correlation of age vs. postgenomic features.
"""
p = OptionParser(correlation.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="12x8")
if len(args) != 1:
sys.exit(not p.print_help())
tsvfile, = args
df = pd.read_csv(tsvfile, sep="\t")
composite_correlation(df, size=(iopts.w, iopts.h))
outfile = tsvfile.rsplit(".", 1)[0] + ".correlation.pdf"
savefig(outfile)
def ccn(args):
"""
%prog ccn combined.tsv
Plot several ccn plots including chr1,chrX,chrY,chrM
"""
p = OptionParser(ccn.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="12x8")
if len(args) != 1:
sys.exit(not p.print_help())
tsvfile, = args
df = pd.read_csv(tsvfile, sep="\t")
composite_ccn(df, size=(iopts.w, iopts.h))
outfile = tsvfile.rsplit(".", 1)[0] + ".ccn.pdf"
savefig(outfile)
def mtdotplots(args):
"""
%prog mtdotplots Mt3.5 Mt4.0 medicago.medicago.lifted.1x1.anchors
Plot Mt3.5 and Mt4.0 side-by-side. This is essentially combined from two
graphics.dotplot() function calls as panel A and B.
"""
from jcvi.graphics.dotplot import check_beds, dotplot
p = OptionParser(mtdotplots.__doc__)
p.set_beds()
opts, args, iopts = p.set_image_options(args, figsize="16x8", dpi=90)
if len(args) != 3:
sys.exit(not p.print_help())
a, b, ac = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
r1 = fig.add_axes([0, 0, .5, 1])
r2 = fig.add_axes([.5, 0, .5, 1])
a1 = fig.add_axes([.05, .1, .4, .8])
a2 = fig.add_axes([.55, .1, .4, .8])
anchorfile = op.join(a, ac)
qbed, sbed, qorder, sorder, is_self = check_beds(anchorfile, p, opts)
dotplot(anchorfile, qbed, sbed, fig, r1, a1, is_self=is_self,
genomenames="Mt3.5_Mt3.5")
opts.qbed = opts.sbed = None
anchorfile = op.join(b, ac)
qbed, sbed, qorder, sorder, is_self = check_beds(anchorfile, p, opts)
dotplot(anchorfile, qbed, sbed, fig, r2, a2, is_self=is_self,
genomenames="Mt4.0_Mt4.0")
root.text(.03, .95, "A", ha="center", va="center", size=36)
root.text(.53, .95, "B", ha="center", va="center", size=36)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "mtdotplots"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def gff(args):
"""
%prog gff *.gff
Draw exons for genes based on gff files. Each gff file should contain only
one gene, and only the "mRNA" and "CDS" feature will be drawn on the canvas.
"""
align_choices = ("left", "center", "right")
p = OptionParser(gff.__doc__)
p.add_option("--align", default="left", choices=align_choices,
help="Horizontal alignment [default: %default]")
p.add_option("--noUTR", default=False, action="store_true",
help="Do not plot UTRs [default: %default]")
opts, args = p.parse_args(args)
if len(args) < 1:
sys.exit(not p.print_help())
fig = plt.figure(1, (8, 5))
root = fig.add_axes([0, 0, 1, 1])
gffiles = args
ngenes = len(gffiles)
canvas = .6
setups, ratio = get_setups(gffiles, canvas=canvas, noUTR=opts.noUTR)
align = opts.align
xs = .2 if align == "left" else .8
yinterval = canvas / ngenes
ys = .8
tip = .01
for genename, mrnabed, cdsbeds in setups:
ExonGlyph(root, xs, ys, mrnabed, cdsbeds, ratio=ratio, align=align)
if align == "left":
root.text(xs - tip, ys, genename, ha="right", va="center")
elif align == "right":
root.text(xs + tip, ys, genename, ha="left", va="center")
ys -= yinterval
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
figname = "exons.pdf"
savefig(figname, dpi=300)
def qc(args):
"""
%prog qc postgenomic-s.tsv
Plot basic statistics of a given sample:
Age, Gender, Ethnicity, Cohort, Chemistry
"""
p = OptionParser(heritability.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="10x6")
if len(args) != 1:
sys.exit(not p.print_help())
tsvfile, = args
df = pd.read_csv(tsvfile, sep="\t")
composite_qc(df, size=(iopts.w, iopts.h))
outfile = tsvfile.rsplit(".", 1)[0] + ".qc.pdf"
savefig(outfile)
def movieframe(args):
"""
%prog movieframe tour test.clm contigs.ref.anchors
Draw heatmap and synteny in the same plot.
"""
p = OptionParser(movieframe.__doc__)
p.add_option("--label", help="Figure title")
p.set_beds()
p.set_outfile(outfile=None)
opts, args, iopts = p.set_image_options(args, figsize="16x8",
style="white", cmap="coolwarm",
format="png", dpi=120)
if len(args) != 3:
sys.exit(not p.print_help())
tour, clmfile, anchorsfile = args
tour = tour.split(",")
image_name = opts.outfile or ("movieframe." + iopts.format)
label = opts.label or op.basename(image_name).rsplit(".", 1)[0]
clm = CLMFile(clmfile)
totalbins, bins, breaks = make_bins(tour, clm.tig_to_size)
M = read_clm(clm, totalbins, bins)
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1]) # whole canvas
ax1 = fig.add_axes([.05, .1, .4, .8]) # heatmap
ax2 = fig.add_axes([.55, .1, .4, .8]) # dot plot
ax2_root = fig.add_axes([.5, 0, .5, 1]) # dot plot canvas
# Left axis: heatmap
plot_heatmap(ax1, M, breaks, iopts)
# Right axis: synteny
qbed, sbed, qorder, sorder, is_self = check_beds(anchorsfile, p, opts,
sorted=False)
dotplot(anchorsfile, qbed, sbed, fig, ax2_root, ax2, sep=False, title="")
root.text(.5, .98, clm.name, color="g", ha="center", va="center")
root.text(.5, .95, label, color="darkslategray", ha="center", va="center")
normalize_axes(root)
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def main():
p = OptionParser(__doc__)
opts, args, iopts = p.set_image_options(figsize="8x7")
if len(args) != 2:
sys.exit(not p.print_help())
seqidsfile, layoutfile = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Karyotype(fig, root, seqidsfile, layoutfile)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "karyotype"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def draw(self, title="*Ks* distribution", filename="Ks_plot.pdf"):
ax = self.ax
ks_max = self.ks_max
lines = self.lines
labels = self.labels
legendp = self.legendp
if len(lines) > 1:
leg = ax.legend(lines, labels, loc=legendp,
shadow=True, fancybox=True, prop={"size": 10})
leg.get_frame().set_alpha(.5)
ax.set_xlim((0, ks_max - self.interval))
ax.set_title(markup(title), fontweight="bold")
ax.set_xlabel(markup('Synonymous substitutions per site (*Ks*)'))
ax.set_ylabel('Percentage of gene pairs')
ax.set_xticklabels(ax.get_xticks(), family='Helvetica')
ax.set_yticklabels(ax.get_yticks(), family='Helvetica')
savefig(filename, dpi=300)
def main():
p = OptionParser(__doc__)
p.add_option(
"--switch",
help="Rename the seqid with two-column file [default: %default]")
p.add_option(
"--tree",
help="Display trees on the bottom of the figure [default: %default]")
p.add_option("--extra", help="Extra features in BED format")
p.add_option(
"--gene_style",
default="Rectangle",
help=
"Default <Rectangle> to plot genes as rectangle. Accept <Arrow> to add orientation of genes."
)
p.add_option(
"--scalebar",
default=False,
action="store_true",
help="Add scale bar to the plot",
)
p.add_option(
"--add_gene_legend",
default=False,
action="store_true",
help="Add forward and reverse strand gene legend to the plot",
)
p.add_option(
"--add_gene_label",
default=False,
action="store_true",
help="Add gene names to the plot",
)
p.add_option(
"--shadestyle",
default="curve",
choices=Shade.Styles,
help="Style of syntenic wedges",
)
opts, args, iopts = p.set_image_options(figsize="8x7")
if len(args) != 3:
sys.exit(not p.print_help())
datafile, bedfile, layoutfile = args
switch = opts.switch
tree = opts.tree
pf = datafile.rsplit(".", 1)[0]
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Synteny(fig,
root,
datafile,
bedfile,
layoutfile,
switch=switch,
tree=tree,
extra_features=opts.extra,
scalebar=opts.scalebar,
shadestyle=opts.shadestyle,
gene_legend=opts.add_gene_legend,
add_gene_label=opts.add_gene_label,
gene_style=opts.gene_style)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def dotplot_main(args):
p = OptionParser(__doc__)
p.set_beds()
p.add_option(
"--synteny",
default=False,
action="store_true",
help="Run a fast synteny scan and display blocks",
)
p.add_option("--cmaptext",
help="Draw colormap box on the bottom-left corner")
p.add_option(
"--vmin",
dest="vmin",
type="float",
default=0,
help="Minimum value in the colormap",
)
p.add_option(
"--vmax",
dest="vmax",
type="float",
default=2,
help="Maximum value in the colormap",
)
p.add_option(
"--nmax",
dest="sample_number",
type="int",
default=10000,
help="Maximum number of data points to plot",
)
p.add_option(
"--minfont",
type="int",
default=4,
help="Do not render labels with size smaller than",
)
p.add_option("--colormap",
help="Two column file, block id to color mapping")
p.add_option(
"--colororientation",
action="store_true",
default=False,
help="Color the blocks based on orientation, similar to mummerplot",
)
p.add_option(
"--nosort",
default=False,
action="store_true",
help="Do not sort the seqids along the axes",
)
p.add_option("--nosep",
default=False,
action="store_true",
help="Do not add contig lines")
p.add_option("--title", help="Title of the dot plot")
p.set_dotplot_opts()
p.set_outfile(outfile=None)
opts, args, iopts = p.set_image_options(args,
figsize="9x9",
style="dark",
dpi=90,
cmap="copper")
if len(args) != 1:
sys.exit(not p.print_help())
(anchorfile, ) = args
qbed, sbed, qorder, sorder, is_self = check_beds(anchorfile,
p,
opts,
sorted=(not opts.nosort))
palette = opts.colormap
if palette:
palette = Palette(palettefile=palette)
elif opts.colororientation:
palette = Palette.from_block_orientation(anchorfile, qbed, sbed)
cmaptext = opts.cmaptext
if anchorfile.endswith(".ks"):
from jcvi.apps.ks import KsFile
logging.debug("Anchors contain Ks values")
cmaptext = cmaptext or "*Ks* values"
anchorksfile = anchorfile + ".anchors"
if need_update(anchorfile, anchorksfile):
ksfile = KsFile(anchorfile)
ksfile.print_to_anchors(anchorksfile)
anchorfile = anchorksfile
if opts.skipempty:
ac = AnchorFile(anchorfile)
if is_self:
qseqids = sseqids = set()
else:
qseqids, sseqids = set(), set()
for pair in ac.iter_pairs():
q, s = pair[:2]
qi, q = qorder[q]
si, s = sorder[s]
qseqids.add(q.seqid)
sseqids.add(s.seqid)
if is_self:
qbed = sbed = subset_bed(qbed, qseqids)
else:
qbed = subset_bed(qbed, qseqids)
sbed = subset_bed(sbed, sseqids)
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1]) # the whole canvas
ax = fig.add_axes([0.1, 0.1, 0.8, 0.8]) # the dot plot
dotplot(
anchorfile,
qbed,
sbed,
fig,
root,
ax,
vmin=opts.vmin,
vmax=opts.vmax,
is_self=is_self,
synteny=opts.synteny,
cmap_text=opts.cmaptext,
cmap=iopts.cmap,
genomenames=opts.genomenames,
sample_number=opts.sample_number,
minfont=opts.minfont,
palette=palette,
sep=(not opts.nosep),
sepcolor=set1[int(opts.theme)],
title=opts.title,
stdpf=(not opts.nostdpf),
chpf=(not opts.nochpf),
)
image_name = opts.outfile or (op.splitext(anchorfile)[0] + "." +
opts.format)
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
fig.clear()
def heatmap(args):
"""
%prog heatmap fastafile chr1
Combine stack plot with heatmap to show abundance of various tracks along
given chromosome. Need to give multiple beds to --stacks and --heatmaps
"""
p = OptionParser(heatmap.__doc__)
p.add_option("--stacks",
default="Exons,Introns,DNA_transposons,Retrotransposons",
help="Features to plot in stackplot [default: %default]")
p.add_option("--heatmaps",
default="Copia,Gypsy,hAT,Helitron,Introns,Exons",
help="Features to plot in heatmaps [default: %default]")
p.add_option("--meres", default=None,
help="Extra centromere / telomere features [default: %default]")
add_window_options(p)
opts, args, iopts = p.set_image_options(args, figsize="8x5")
if len(args) != 2:
sys.exit(not p.print_help())
fastafile, chr = args
window, shift, subtract = check_window_options(opts)
stacks = opts.stacks.split(",")
heatmaps = opts.heatmaps.split(",")
stackbeds = get_beds(stacks)
heatmapbeds = get_beds(heatmaps)
stackbins = get_binfiles(stackbeds, fastafile, shift, subtract=subtract)
heatmapbins = get_binfiles(heatmapbeds, fastafile, shift, subtract=subtract)
margin = .06
inner = .015
clen = Sizes(fastafile).mapping[chr]
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
# Gauge
ratio = draw_gauge(root, margin, clen, rightmargin=4 * margin)
yinterval = .3
xx = margin
yy = 1 - margin
yy -= yinterval
xlen = clen / ratio
cc = chr
if "_" in chr:
ca, cb = chr.split("_")
cc = ca[0].upper() + cb
root.add_patch(Rectangle((xx, yy), xlen, yinterval - inner, color=gray))
ax = fig.add_axes([xx, yy, xlen, yinterval - inner])
nbins = get_nbins(clen, shift)
owindow = clen / 100
if owindow > window:
window = owindow / shift * shift
stackplot(ax, stackbins, nbins, palette, chr, window, shift)
ax.text(.1, .9, cc, va="top", zorder=100, transform=ax.transAxes,
bbox=dict(boxstyle="round", fc="w", alpha=.5))
# Legends
xx += xlen + .01
yspace = (yinterval - inner) / (len(stackbins) + 1)
yy = 1 - margin - yinterval
for s, p in zip(stacks, palette):
s = s.replace("_", " ")
s = Registration.get(s, s)
yy += yspace
root.add_patch(Rectangle((xx, yy), inner, inner, color=p, lw=0))
root.text(xx + 1.5 * inner, yy, s, size=10)
yh = .05 # Heatmap height
# Heatmaps
xx = margin
yy = 1 - margin - yinterval - inner
for s, p in zip(heatmaps, heatmapbins):
s = s.replace("_", " ")
s = Registration.get(s, s)
yy -= yh
m = stackarray(p, chr, window, shift)
Y = np.array([m, m])
root.imshow(Y, extent=(xx, xx + xlen, yy, yy + yh - inner),
interpolation="nearest", aspect="auto")
root.text(xx + xlen + .01, yy, s, size=10)
yy -= yh
meres = opts.meres
if meres:
bed = Bed(meres)
for b in bed:
if b.seqid != chr:
continue
pos = (b.start + b.end) / 2
cpos = pos / ratio
xx = margin + cpos
accn = b.accn.capitalize()
root.add_patch(CirclePolygon((xx, yy), radius=.01, fc="m", ec="m"))
root.text(xx + .014, yy, accn, va="center", color="m")
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = chr + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def main():
p = OptionParser(__doc__)
p.add_option("--switch", help="Rename the seqid with two-column file")
p.add_option("--tree", help="Display trees on the bottom of the figure")
p.add_option("--extra", help="Extra features in BED format")
p.add_option(
"--genelabelsize",
default=0,
type="int",
help="Show gene labels at this font size, useful for debugging. " +
"However, plot may appear visually crowded. " +
"Reasonably good values are 2 to 6 [Default: disabled]",
)
p.add_option(
"--scalebar",
default=False,
action="store_true",
help="Add scale bar to the plot",
)
p.add_option(
"--glyphstyle",
default="box",
choices=Glyph.Styles,
help="Style of feature glyphs",
)
p.add_option(
"--glyphcolor",
default="orientation",
choices=Glyph.Palette,
help="Glyph coloring based on",
)
p.add_option(
"--shadestyle",
default="curve",
choices=Shade.Styles,
help="Style of syntenic wedges",
)
opts, args, iopts = p.set_image_options(figsize="8x7")
if len(args) != 3:
sys.exit(not p.print_help())
datafile, bedfile, layoutfile = args
switch = opts.switch
tree = opts.tree
pf = datafile.rsplit(".", 1)[0]
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Synteny(
fig,
root,
datafile,
bedfile,
layoutfile,
switch=switch,
tree=tree,
extra_features=opts.extra,
genelabelsize=opts.genelabelsize,
scalebar=opts.scalebar,
shadestyle=opts.shadestyle,
glyphstyle=opts.glyphstyle,
glyphcolor=opts.glyphcolor,
)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def wheel(args):
"""
%prog wheel datafile.csv groups.csv
Wheel plot that shows continous data in radial axes.
"""
p = OptionParser(wheel.__doc__)
p.add_option(
"--column",
default="score",
choices=("score", "percentile"),
help="Which column to extract from `datafile.csv`",
)
opts, args, iopts = p.set_image_options(args, figsize="5x5", format="png")
if len(args) != 2:
sys.exit(not p.print_help())
datafile, groupsfile = args
column = opts.column
linecolor = "#d6d6d6"
df = parse_data(datafile, score_column=opts.column)
groups = parse_groups(groupsfile)
labels = [g for g in groups if g in df]
print(labels)
df = [df[g] for g in labels]
print(df)
groups = [groups[g] for g in labels]
print(groups)
pf = datafile.rsplit(".", 1)[0]
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
categories = len(df)
# ax = plt.subplot(111, projection='polar')
ax = fig.add_axes([0.1, 0.1, 0.8, 0.8], polar=True)
brewer = [
"#FF3B30",
"#DD43A0",
"#5856D6",
"#007AFE",
"#56BDEC",
"#4CD8BA",
"#4CD864",
"#B0F457",
"#FEF221",
"#FFCC01",
"#FF9500",
"#FF3B30",
]
# Baseline
theta = np.linspace(1.5 * np.pi,
3.5 * np.pi,
endpoint=False,
num=categories)
_theta = np.linspace(1.5 * np.pi, 3.5 * np.pi)
R = max(max(df), 10)
xlim = (-R, R) if column == "score" else (-100, 100)
plim = (-R / 2, R) if column == "score" else (0, 100)
ci = (-0.5, 2) if column == "score" else (10, 90)
# Grid
if column == "score":
for t in theta:
ax.plot([t, t], plim, color=linecolor)
ax.axis("off")
# Contours
for t in plim:
ax.plot(_theta, [t] * len(_theta), color=linecolor)
# Sectors (groupings)
collapsed_groups = []
gg = []
for group, c in groupby(enumerate(groups), lambda x: x[1]):
c = [x[0] for x in list(c)]
collapsed_groups.append(group)
gg.append(c)
show_sector = False
if show_sector:
theta_interval = 2 * np.pi / categories
theta_pad = theta_interval / 2 * 0.9
for color, group in zip(brewer, gg):
tmin, tmax = min(group), max(group)
sector(
ax,
theta[tmin],
theta[tmax],
theta_pad,
R * 0.95,
ls="-",
color=color,
lw=2,
)
# Data
r = df
closed_plot(ax, theta, r, color="lightslategray", alpha=0.25)
for color, group in zip(brewer, gg):
hidden_data = [(theta[x], r[x]) for x in group
if (ci[0] <= r[x] <= ci[1])]
shown_data = [(theta[x], r[x]) for x in group
if (r[x] < ci[0] or r[x] > ci[1])]
for alpha, data in zip((1, 1), (hidden_data, shown_data)):
if not data:
continue
color_theta, color_r = zip(*data)
ax.plot(color_theta, color_r, "o", color=color, alpha=alpha)
# Print out data
diseaseNames, risks = labels, df
print("var theta = [{}]".format(",".join("{:.1f}".format(degrees(x))
for x in theta)))
print("var risks = [{}]".format(",".join(str(x) for x in risks)))
print("var diseaseNames = [{}]".format(",".join(
['"{}"'.format(x) for x in diseaseNames])))
# Labels
from math import cos, sin
r = 0.5
for i, label in enumerate(labels):
tl = theta[i]
x, y = 0.5 + r * cos(tl), 0.5 + r * sin(tl)
d = degrees(tl)
if 90 < d % 360 < 270: # On the left quardrants
d -= 180
root.text(x,
y,
label,
size=4,
rotation=d,
ha="center",
va="center",
color=linecolor)
print(x, y, label)
# Add baseline
baseline = 0 if column == "score" else 50
_r = len(_theta) * [baseline]
closed_plot(ax, _theta, _r, "k:", lw=1, ms=4)
# Add confidence interval
if column == "percentile":
barcolor = "#eeeeee"
ax.bar([0], [ci[1] - ci[0]],
width=2 * np.pi,
bottom=ci[0],
fc=barcolor)
ax.set_rmin(xlim[0])
ax.set_rmax(xlim[1])
normalize_axes(root)
image_name = pf + "-" + column + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def main(args):
"""
%prog newicktree
Plot Newick formatted tree. The gene structure can be plotted along if
--gffdir is given. The gff file needs to be `genename.gff`. If --sizes is
on, also show the number of amino acids.
With --barcode a mapping file can be provided to convert seq names to
eg. species names, useful in unified tree display. This file should have
distinctive barcodes in column1 and new names in column2, tab delimited.
"""
p = OptionParser(main.__doc__)
p.add_option("--outgroup", help="Outgroup for rerooting the tree. " + \
"Use comma to separate multiple taxa.")
p.add_option("--noreroot", default=False, action="store_true", \
help="Don't reroot the input tree [default: %default]")
p.add_option("--rmargin", default=.3, type="float",
help="Set blank rmargin to the right [default: %default]")
p.add_option("--gffdir", default=None,
help="The directory that contain GFF files [default: %default]")
p.add_option("--sizes", default=None,
help="The FASTA file or the sizes file [default: %default]")
p.add_option("--SH", default=None, type="string",
help="SH test p-value [default: %default]")
p.add_option("--scutoff", default=0, type="int",
help="cutoff for displaying node support, 0-100 [default: %default]")
p.add_option("--barcode", default=None,
help="path to seq names barcode mapping file: " \
"barcode<tab>new_name [default: %default]")
p.add_option("--leafcolor", default="k",
help="Font color for the OTUs, or path to a file " \
"containing color mappings: leafname<tab>color [default: %default]")
p.add_option("--leaffont", default=12, help="Font size for the OTUs")
p.add_option("--geoscale", default=False, action="store_true",
help="Plot geological scale")
opts, args, iopts = p.set_image_options(args, figsize="8x6")
if len(args) != 1:
sys.exit(not p.print_help())
datafile, = args
outgroup = None
reroot = not opts.noreroot
if opts.outgroup:
outgroup = opts.outgroup.split(",")
if datafile == "demo":
tx = """(((Os02g0681100:0.1151,Sb04g031800:0.11220)1.0:0.0537,
(Os04g0578800:0.04318,Sb06g026210:0.04798)-1.0:0.08870)1.0:0.06985,
((Os03g0124100:0.08845,Sb01g048930:0.09055)1.0:0.05332,
(Os10g0534700:0.06592,Sb01g030630:0.04824)-1.0:0.07886):0.09389);"""
else:
logging.debug("Load tree file `{0}`.".format(datafile))
tx = open(datafile).read()
pf = datafile.rsplit(".", 1)[0]
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
if opts.geoscale:
draw_geoscale(root)
else:
if op.isfile(opts.leafcolor):
leafcolor = "k"
leafcolorfile = opts.leafcolor
else:
leafcolor = opts.leafcolor
leafcolorfile = None
draw_tree(root, tx, rmargin=opts.rmargin, leafcolor=leafcolor, \
outgroup=outgroup, reroot=reroot, gffdir=opts.gffdir, \
sizes=opts.sizes, SH=opts.SH, scutoff=opts.scutoff, \
barcodefile=opts.barcode, leafcolorfile=leafcolorfile,
leaffont=opts.leaffont)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def lms(args):
"""
%prog lms
ALLMAPS cartoon to illustrate LMS metric.
"""
from random import randint
from jcvi.graphics.chromosome import HorizontalChromosome
p = OptionParser(lms.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="6x6", dpi=300)
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
# Panel A
w, h = .7, .35
ax = fig.add_axes([.15, .6, w, h])
xdata = [x + randint(-3, 3) for x in range(10, 110, 10)]
ydata = [x + randint(-3, 3) for x in range(10, 110, 10)]
ydata[3:7] = ydata[3:7][::-1]
xydata = zip(xdata, ydata)
lis = xydata[:3] + [xydata[4]] + xydata[7:]
lds = xydata[3:7]
xlis, ylis = zip(*lis)
xlds, ylds = zip(*lds)
ax.plot(xlis,
ylis,
"r-",
lw=12,
alpha=.3,
solid_capstyle="round",
solid_joinstyle="round")
ax.plot(xlds,
ylds,
"g-",
lw=12,
alpha=.3,
solid_capstyle="round",
solid_joinstyle="round")
ax.plot(xdata, ydata, "k.", mec="k", mfc="w", mew=3, ms=12)
HorizontalChromosome(root, .15, .15 + w, .57, height=.02, lw=2)
root.text(.15 + w / 2,
.55,
"Chromosome location (bp)",
ha="center",
va="top")
ax.text(80, 30, "LIS = 7", color="r", ha="center", va="center")
ax.text(80, 20, "LDS = 4", color="g", ha="center", va="center")
ax.text(80, 10, "LMS = $max$(LIS, LDS) = 7", ha="center", va="center")
normalize_lms_axis(ax)
# Panel B
w = .37
p = (0, 45, 75, 110)
ax = fig.add_axes([.1, .12, w, h])
xdata = [x for x in range(10, 110, 10)]
ydata = ydata_orig = [x for x in range(10, 110, 10)]
ydata = ydata[:4] + ydata[7:] + ydata[4:7][::-1]
xydata = zip(xdata, ydata)
lis = xydata[:7]
xlis, ylis = zip(*lis)
ax.plot(xlis,
ylis,
"r-",
lw=12,
alpha=.3,
solid_capstyle="round",
solid_joinstyle="round")
ax.plot(xdata, ydata, "k.", mec="k", mfc="w", mew=3, ms=12)
ax.vlines(p, 0, 110, colors="beige", lw=3)
normalize_lms_axis(ax)
patch = [.1 + w * x / 110. for x in p]
HorizontalChromosome(root, .1, .1 + w, .09, patch=patch, height=.02, lw=2)
scaffolds = ("a", "b", "c")
for i, s in enumerate(scaffolds):
xx = (patch[i] + patch[i + 1]) / 2
root.text(xx, .09, s, va="center", ha="center")
root.text(.1 + w / 2, .04, "LMS($a||b||c$) = 7", ha="center")
# Panel C
ax = fig.add_axes([.6, .12, w, h])
patch = [.6 + w * x / 110. for x in p]
ydata = ydata_orig
ax.plot(xdata,
ydata,
"r-",
lw=12,
alpha=.3,
solid_capstyle="round",
solid_joinstyle="round")
ax.plot(xdata, ydata, "k.", mec="k", mfc="w", mew=3, ms=12)
ax.vlines(p, [0], [110], colors="beige", lw=3)
normalize_lms_axis(ax)
HorizontalChromosome(root, .6, .6 + w, .09, patch=patch, height=.02, lw=2)
scaffolds = ("a", "-c", "b")
for i, s in enumerate(scaffolds):
xx = (patch[i] + patch[i + 1]) / 2
root.text(xx, .09, s, va="center", ha="center")
root.text(.6 + w / 2, .04, "LMS($a||-c||b$) = 10", ha="center")
labels = ((.05, .95, 'A'), (.05, .48, 'B'), (.55, .48, 'C'))
panel_labels(root, labels)
normalize_axes(root)
pf = "lms"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def covlen(args):
"""
%prog covlen covfile fastafile
Plot coverage vs length. `covfile` is two-column listing contig id and
depth of coverage.
"""
import numpy as np
import pandas as pd
import seaborn as sns
from jcvi.formats.base import DictFile
p = OptionParser(covlen.__doc__)
p.add_option("--maxsize",
default=1000000,
type="int",
help="Max contig size")
p.add_option("--maxcov", default=100, type="int", help="Max contig size")
p.add_option("--color", default='m', help="Color of the data points")
p.add_option("--kind",
default="scatter",
choices=("scatter", "reg", "resid", "kde", "hex"),
help="Kind of plot to draw")
opts, args, iopts = p.set_image_options(args, figsize="8x8")
if len(args) != 2:
sys.exit(not p.print_help())
covfile, fastafile = args
cov = DictFile(covfile, cast=float)
s = Sizes(fastafile)
data = []
maxsize, maxcov = opts.maxsize, opts.maxcov
for ctg, size in s.iter_sizes():
c = cov.get(ctg, 0)
if size > maxsize:
continue
if c > maxcov:
continue
data.append((size, c))
x, y = zip(*data)
x = np.array(x)
y = np.array(y)
logging.debug("X size {0}, Y size {1}".format(x.size, y.size))
df = pd.DataFrame()
xlab, ylab = "Length", "Coverage of depth (X)"
df[xlab] = x
df[ylab] = y
sns.jointplot(xlab,
ylab,
kind=opts.kind,
data=df,
xlim=(0, maxsize),
ylim=(0, maxcov),
stat_func=None,
edgecolor="w",
color=opts.color)
figname = covfile + ".pdf"
savefig(figname, dpi=iopts.dpi, iopts=iopts)
def coverage(args):
"""
%prog coverage fastafile ctg bedfile1 bedfile2 ..
Plot coverage from a set of BED files that contain the read mappings. The
paired read span will be converted to a new bedfile that contain the happy
mates. ctg is the chr/scf/ctg that you want to plot the histogram on.
If the bedfiles already contain the clone spans, turn on --spans.
"""
from jcvi.formats.bed import mates, bedpe
p = OptionParser(coverage.__doc__)
p.add_option("--ymax",
default=None,
type="int",
help="Limit ymax [default: %default]")
p.add_option(
"--spans",
default=False,
action="store_true",
help="BED files already contain clone spans [default: %default]")
opts, args, iopts = p.set_image_options(args, figsize="8x5")
if len(args) < 3:
sys.exit(not p.print_help())
fastafile, ctg = args[0:2]
bedfiles = args[2:]
sizes = Sizes(fastafile)
size = sizes.mapping[ctg]
plt.figure(1, (iopts.w, iopts.h))
ax = plt.gca()
bins = 100 # smooth the curve
lines = []
legends = []
not_covered = []
yy = .9
for bedfile, c in zip(bedfiles, "rgbcky"):
if not opts.spans:
pf = bedfile.rsplit(".", 1)[0]
matesfile = pf + ".mates"
if need_update(bedfile, matesfile):
matesfile, matesbedfile = mates([bedfile, "--lib"])
bedspanfile = pf + ".spans.bed"
if need_update(matesfile, bedspanfile):
bedpefile, bedspanfile = bedpe(
[bedfile, "--span", "--mates={0}".format(matesfile)])
bedfile = bedspanfile
bedsum = Bed(bedfile).sum(seqid=ctg)
notcoveredbases = size - bedsum
legend = bedfile.split(".")[0]
msg = "{0}: {1} bp not covered".format(legend,
thousands(notcoveredbases))
not_covered.append(msg)
print(msg, file=sys.stderr)
ax.text(.1, yy, msg, color=c, size=9, transform=ax.transAxes)
yy -= .08
cov = Coverage(bedfile, sizes.filename)
x, y = cov.get_plot_data(ctg, bins=bins)
line, = ax.plot(x, y, '-', color=c, lw=2, alpha=.5)
lines.append(line)
legends.append(legend)
leg = ax.legend(lines, legends, shadow=True, fancybox=True)
leg.get_frame().set_alpha(.5)
ylabel = "Average depth per {0}Kb".format(size / bins / 1000)
ax.set_xlim(0, size)
ax.set_ylim(0, opts.ymax)
ax.set_xlabel(ctg)
ax.set_ylabel(ylabel)
set_human_base_axis(ax)
figname = "{0}.{1}.pdf".format(fastafile, ctg)
savefig(figname, dpi=iopts.dpi, iopts=iopts)
def astat(args):
"""
%prog astat coverage.log
Create coverage-rho scatter plot.
"""
p = OptionParser(astat.__doc__)
p.add_option("--cutoff", default=1000, type="int", help="Length cutoff")
p.add_option("--genome", default="", help="Genome name")
p.add_option(
"--arrDist",
default=False,
action="store_true",
help="Use arrDist instead",
)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
(covfile, ) = args
cutoff = opts.cutoff
genome = opts.genome
plot_arrDist = opts.arrDist
suffix = ".{0}".format(cutoff)
small_covfile = covfile + suffix
update_covfile = need_update(covfile, small_covfile)
if update_covfile:
fw = open(small_covfile, "w")
else:
logging.debug("Found `{0}`, will use this one".format(small_covfile))
covfile = small_covfile
fp = open(covfile)
header = next(fp)
if update_covfile:
fw.write(header)
data = []
msg = "{0} tigs scanned ..."
for row in fp:
tigID, rho, covStat, arrDist = row.split()
tigID = int(tigID)
if tigID % 1000000 == 0:
sys.stderr.write(msg.format(tigID) + "\r")
rho, covStat, arrDist = [float(x) for x in (rho, covStat, arrDist)]
if rho < cutoff:
continue
if update_covfile:
fw.write(row)
data.append((tigID, rho, covStat, arrDist))
print(msg.format(tigID), file=sys.stderr)
from jcvi.graphics.base import plt, savefig
logging.debug("Plotting {0} data points.".format(len(data)))
tigID, rho, covStat, arrDist = zip(*data)
y = arrDist if plot_arrDist else covStat
ytag = "arrDist" if plot_arrDist else "covStat"
fig = plt.figure(1, (7, 7))
ax = fig.add_axes([0.12, 0.1, 0.8, 0.8])
ax.plot(rho, y, ".", color="lightslategrey")
xtag = "rho"
info = (genome, xtag, ytag)
title = "{0} {1} vs. {2}".format(*info)
ax.set_title(title)
ax.set_xlabel(xtag)
ax.set_ylabel(ytag)
if plot_arrDist:
ax.set_yscale("log")
imagename = "{0}.png".format(".".join(info))
savefig(imagename, dpi=150)
def depth(args):
"""
%prog depth *.regions.bed.gz
Plot the mosdepth regions BED file. We recommend to generate this BED file
by (please adjust the --by parameter to your required resolution):
$ mosdepth --no-per-base --use-median --fast-mode --by 1000000 sample.wgs
sample.bam
Use --chrinfo to specify a colormap between seqid, desired color, and
optionally a new name. For example:
chr01A, #c51b7d, 1A
chr01B, #4d9221, 1B
...
Only seqids that are in the colormap will be plotted, in the order that's
given in the file. When --colormap is not set, every seqid will be drawn in
black.
Can take multiple BED files as input and then plot all of them in a
composite figure.
"""
p = OptionParser(depth.__doc__)
p.add_option(
"--chrinfo",
help="Comma-separated mappings between seqid, color, new_name")
p.add_option(
"--titleinfo",
help="Comma-separated titles mappings between filename, title",
)
opts, args, iopts = p.set_image_options(args, style="dark", figsize="14x4")
if len(args) < 1:
sys.exit(not p.print_help())
bedfiles = args
chrinfo = ChrInfoFile(opts.chrinfo) if opts.chrinfo else {}
titleinfo = TitleInfoFile(opts.titleinfo) if opts.titleinfo else {}
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
npanels = len(bedfiles)
yinterval = 1.0 / npanels
ypos = 1 - yinterval
for bedfile in bedfiles:
pf = op.basename(bedfile).split(".", 1)[0]
bed = Bed(bedfile)
panel_root = root if npanels == 1 else fig.add_axes(
[0, ypos, 1, yinterval])
panel_ax = fig.add_axes(
[0.1, ypos + 0.2 * yinterval, 0.8, 0.65 * yinterval])
if ypos > 0.001:
root.plot((0, 1), (ypos, ypos), "-", lw=2, color="lightslategray")
title = titleinfo.get(bedfile, pf.split("_", 1)[0])
subtitle = None
if isinstance(title, TitleInfoLine):
subtitle = title.subtitle
title = title.title
draw_depth(panel_root,
panel_ax,
bed,
chrinfo=chrinfo,
title=title,
subtitle=subtitle)
ypos -= yinterval
normalize_axes(root)
if npanels > 1:
pf = op.commonprefix(bedfiles)
pf = pf or "depth"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def simulate(args):
"""
%prog simulate
Run simulation on female restitution.
"""
import seaborn as sns
sns.set_style("darkgrid")
p = OptionParser(simulate.__doc__)
p.add_option(
"--verbose",
default=False,
action="store_true",
help="Verbose logging during simulation",
)
opts, args, iopts = p.set_image_options(args, figsize="6x6")
if len(args) != 0:
sys.exit(not p.print_help())
# Construct a composite figure with 6 tracks
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
rows = 6
ypad = 0.05
yinterval = (1 - 2 * ypad) / (rows + 1)
yy = 1 - ypad
xpad = 0.2
xwidth = 0.7
# Axes are vertically stacked, and share x-axis
axes = []
yy_positions = [
] # Save yy positions so we can show details to the right laterr
for idx in range(rows):
yy_positions.append(yy)
yy -= yinterval
ax = fig.add_axes([xpad, yy, xwidth, yinterval * 0.85])
if idx != rows - 1:
plt.setp(ax.get_xticklabels(), visible=False)
axes.append(ax)
ax1, ax2, ax3, ax4, ax5, ax6 = axes
# Prepare the simulated data
# Simulate two parents
SS = Genome("SS", "SS", 10, 8)
SO = Genome("SO", "SO", 8, 10)
verbose = opts.verbose
all_F1s = [simulate_F1(SO, SS, verbose=verbose) for _ in range(1000)]
all_F2s = [simulate_F2(SO, SS, verbose=verbose) for _ in range(1000)]
all_BC1s = [simulate_BCn(1, SO, SS, verbose=verbose) for _ in range(1000)]
all_BC2s = [simulate_BCn(2, SO, SS, verbose=verbose) for _ in range(1000)]
all_BC3s = [simulate_BCn(3, SO, SS, verbose=verbose) for _ in range(1000)]
all_BC4s = [simulate_BCn(4, SO, SS, verbose=verbose) for _ in range(1000)]
# Plotting
plot_summary(ax1, all_F1s)
plot_summary(ax2, all_F2s)
plot_summary(ax3, all_BC1s)
plot_summary(ax4, all_BC2s)
plot_summary(ax5, all_BC3s)
plot_summary(ax6, all_BC4s)
# Show title to the left
xx = xpad / 2
for (title, subtitle), yy in zip(
(
("F1", None),
("F2", None),
("BC1", None),
("BC2", None),
("BC3", None),
("BC4", None),
),
yy_positions,
):
if subtitle:
yy -= 0.06
else:
yy -= 0.07
root.text(
xx,
yy,
title,
color="darkslategray",
ha="center",
va="center",
fontweight="semibold",
)
if subtitle:
yy -= 0.02
root.text(xx,
yy,
subtitle,
color="lightslategray",
ha="center",
va="center")
axes[-1].set_xlabel("Number of unique chromosomes")
adjust_spines(axes[-1], ["bottom"], outward=True)
normalize_axes(root)
savefig("plotter.pdf", dpi=120)
outdir = "simulations"
mkdir(outdir)
# Write chromosomes to disk
for genomes, filename in (
(all_F1s, "all_F1s"),
(all_F2s, "all_F2s"),
(all_BC1s, "all_BC1s"),
(all_BC2s, "all_BC2s"),
(all_BC3s, "all_BC3s"),
(all_BC4s, "all_BC4s"),
):
write_chromosomes(genomes, op.join(outdir, filename))
def plot(args):
"""
%prog plot input.bed seqid
Plot the matchings between the reconstructed pseudomolecules and the maps.
Two types of visualizations are available in one canvas:
1. Parallel axes, and matching markers are shown in connecting lines;
2. Scatter plot.
"""
from jcvi.graphics.base import plt, savefig, normalize_axes, \
set2, panel_labels
from jcvi.graphics.chromosome import Chromosome, GeneticMap, \
HorizontalChromosome
p = OptionParser(plot.__doc__)
add_allmaps_plot_options(p)
opts, args, iopts = p.set_image_options(args, figsize="10x6")
if len(args) != 2:
sys.exit(not p.print_help())
inputbed, seqid = args
pf = inputbed.rsplit(".", 1)[0]
bedfile = pf + ".lifted.bed"
agpfile = pf + ".agp"
weightsfile = opts.weightsfile
links = opts.links
function = get_function(opts.distance)
cc = Map(bedfile, function)
allseqids = cc.seqids
mapnames = cc.mapnames
weights = Weights(weightsfile, mapnames)
assert seqid in allseqids, "{0} not in {1}".format(seqid, allseqids)
s = Scaffold(seqid, cc)
mlgs = [k for k, v in s.mlg_counts.items() if v >= links]
while not mlgs:
links /= 2
logging.error("No markers to plot, --links reset to {0}".format(links))
mlgs = [k for k, v in s.mlg_counts.items() if v >= links]
mlgsizes = {}
for mlg in mlgs:
mm = cc.extract_mlg(mlg)
mlgsize = max(function(x) for x in mm)
mlgsizes[mlg] = mlgsize
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
ax1 = fig.add_axes([0, 0, .5, 1])
ax2 = fig.add_axes([.5, 0, .5, 1])
# Find the layout first
ystart, ystop = .9, .1
L = Layout(mlgsizes)
coords = L.coords
tip = .02
marker_pos = {}
# Palette
colors = dict((mapname, set2[i]) for i, mapname in enumerate(mapnames))
colors = dict((mlg, colors[mlg.split("-")[0]]) for mlg in mlgs)
rhos = {}
# Parallel coordinates
for mlg, (x, y1, y2) in coords.items():
mm = cc.extract_mlg(mlg)
markers = [(m.accn, function(m)) for m in mm] # exhaustive marker list
xy = [(m.pos, function(m)) for m in mm if m.seqid == seqid]
mx, my = zip(*xy)
rho = spearmanr(mx, my)
rhos[mlg] = rho
flip = rho < 0
g = GeneticMap(ax1, x, y1, y2, markers, tip=tip, flip=flip)
extra = -3 * tip if x < .5 else 3 * tip
ha = "right" if x < .5 else "left"
mapname = mlg.split("-")[0]
tlg = mlg.replace("_", ".") # Latex does not like underscore char
label = "{0} (w={1})".format(tlg, weights[mapname])
ax1.text(x + extra, (y1 + y2) / 2, label, color=colors[mlg],
ha=ha, va="center", rotation=90)
marker_pos.update(g.marker_pos)
agp = AGP(agpfile)
agp = [x for x in agp if x.object == seqid]
chrsize = max(x.object_end for x in agp)
# Pseudomolecules in the center
r = ystart - ystop
ratio = r / chrsize
f = lambda x: (ystart - ratio * x)
patchstart = [f(x.object_beg) for x in agp if not x.is_gap]
Chromosome(ax1, .5, ystart, ystop, width=2 * tip, patch=patchstart, lw=2)
label = "{0} ({1})".format(seqid, human_size(chrsize, precision=0))
ax1.text(.5, ystart + tip, label, ha="center")
scatter_data = defaultdict(list)
# Connecting lines
for b in s.markers:
marker_name = b.accn
if marker_name not in marker_pos:
continue
cx = .5
cy = f(b.pos)
mx = coords[b.mlg][0]
my = marker_pos[marker_name]
extra = -tip if mx < cx else tip
extra *= 1.25 # leave boundaries for aesthetic reasons
cx += extra
mx -= extra
ax1.plot((cx, mx), (cy, my), "-", color=colors[b.mlg])
scatter_data[b.mlg].append((b.pos, function(b)))
# Scatter plot, same data as parallel coordinates
xstart, xstop = sorted((ystart, ystop))
f = lambda x: (xstart + ratio * x)
pp = [x.object_beg for x in agp if not x.is_gap]
patchstart = [f(x) for x in pp]
HorizontalChromosome(ax2, xstart, xstop, ystop,
height=2 * tip, patch=patchstart, lw=2)
gap = .03
ratio = (r - gap * len(mlgs) - tip) / sum(mlgsizes.values())
tlgs = []
for mlg, mlgsize in sorted(mlgsizes.items()):
height = ratio * mlgsize
ystart -= height
xx = .5 + xstart / 2
width = r / 2
color = colors[mlg]
ax = fig.add_axes([xx, ystart, width, height])
ypos = ystart + height / 2
ystart -= gap
sd = scatter_data[mlg]
xx, yy = zip(*sd)
ax.vlines(pp, 0, mlgsize, colors="beige")
ax.plot(xx, yy, ".", color=color)
rho = rhos[mlg]
ax.text(.5, 1 - .4 * gap / height, r"$\rho$={0:.3f}".format(rho),
ha="center", va="top", transform=ax.transAxes, color="gray")
tlg = mlg.replace("_", ".")
tlgs.append((tlg, ypos, color))
ax.set_xlim(0, chrsize)
ax.set_ylim(0, mlgsize)
ax.set_xticks([])
while height / len(ax.get_yticks()) < .03 and len(ax.get_yticks()) >= 2:
ax.set_yticks(ax.get_yticks()[::2]) # Sparsify the ticks
yticklabels = [int(x) for x in ax.get_yticks()]
ax.set_yticklabels(yticklabels, family='Helvetica')
if rho < 0:
ax.invert_yaxis()
for i, (tlg, ypos, color) in enumerate(tlgs):
ha = "center"
if len(tlgs) > 4:
ha = "right" if i % 2 else "left"
root.text(.5, ypos, tlg, color=color, rotation=90,
ha=ha, va="center")
if opts.panels:
labels = ((.04, .96, 'A'), (.48, .96, 'B'))
panel_labels(root, labels)
normalize_axes((ax1, ax2, root))
image_name = seqid + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
plt.close(fig)
def histogram(args):
"""
%prog histogram meryl.histogram species K
Plot the histogram based on meryl K-mer distribution, species and N are
only used to annotate the graphic.
"""
p = OptionParser(histogram.__doc__)
p.add_option("--vmin", dest="vmin", default=1, type="int",
help="minimum value, inclusive [default: %default]")
p.add_option("--vmax", dest="vmax", default=100, type="int",
help="maximum value, inclusive [default: %default]")
p.add_option("--pdf", default=False, action="store_true",
help="Print PDF instead of ASCII plot [default: %default]")
p.add_option("--coverage", default=0, type="int",
help="Kmer coverage [default: auto]")
p.add_option("--nopeaks", default=False, action="store_true",
help="Do not annotate K-mer peaks")
opts, args = p.parse_args(args)
if len(args) != 3:
sys.exit(not p.print_help())
histfile, species, N = args
ascii = not opts.pdf
peaks = not opts.nopeaks
N = int(N)
if histfile.rsplit(".", 1)[-1] in ("mcdat", "mcidx"):
logging.debug("CA kmer index found")
histfile = merylhistogram(histfile)
ks = KmerSpectrum(histfile)
ks.analyze(K=N)
Total_Kmers = int(ks.totalKmers)
coverage = opts.coverage
Kmer_coverage = ks.max2 if not coverage else coverage
Genome_size = int(round(Total_Kmers * 1. / Kmer_coverage))
Total_Kmers_msg = "Total {0}-mers: {1}".format(N, thousands(Total_Kmers))
Kmer_coverage_msg = "{0}-mer coverage: {1}".format(N, Kmer_coverage)
Genome_size_msg = "Estimated genome size: {0:.1f}Mb".\
format(Genome_size / 1e6)
Repetitive_msg = ks.repetitive
SNPrate_msg = ks.snprate
for msg in (Total_Kmers_msg, Kmer_coverage_msg, Genome_size_msg):
print >> sys.stderr, msg
x, y = ks.get_xy(opts.vmin, opts.vmax)
title = "{0} {1}-mer histogram".format(species, N)
if ascii:
asciiplot(x, y, title=title)
return Genome_size
plt.figure(1, (6, 6))
plt.plot(x, y, 'g-', lw=2, alpha=.5)
ax = plt.gca()
if peaks:
t = (ks.min1, ks.max1, ks.min2, ks.max2, ks.min3)
tcounts = [(x, y) for x, y in ks.counts if x in t]
if tcounts:
x, y = zip(*tcounts)
tcounts = dict(tcounts)
plt.plot(x, y, 'ko', lw=2, mec='k', mfc='w')
ax.text(ks.max1, tcounts[ks.max1], "SNP peak", va="top")
ax.text(ks.max2, tcounts[ks.max2], "Main peak")
messages = [Total_Kmers_msg, Kmer_coverage_msg, Genome_size_msg,
Repetitive_msg, SNPrate_msg]
write_messages(ax, messages)
ymin, ymax = ax.get_ylim()
ymax = ymax * 7 / 6
ax.set_title(markup(title))
ax.set_ylim((ymin, ymax))
xlabel, ylabel = "Coverage (X)", "Counts"
ax.set_xlabel(xlabel)
ax.set_ylabel(ylabel)
set_human_axis(ax)
imagename = histfile.split(".")[0] + ".pdf"
savefig(imagename, dpi=100)
return Genome_size
def estimategaps(args):
"""
%prog estimategaps JM-4 chr1 JMMale-1
Illustrate ALLMAPS gap estimation algorithm.
"""
p = OptionParser(estimategaps.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="6x6", dpi=300)
if len(args) != 3:
sys.exit(not p.print_help())
pf, seqid, mlg = args
bedfile = pf + ".lifted.bed"
agpfile = pf + ".agp"
function = lambda x: x.cm
cc = Map(bedfile, scaffold_info=True, function=function)
agp = AGP(agpfile)
g = GapEstimator(cc, agp, seqid, mlg, function=function)
pp, chrsize, mlgsize = g.pp, g.chrsize, g.mlgsize
spl, spld = g.spl, g.spld
g.compute_all_gaps(verbose=False)
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
# Panel A
xstart, ystart = .15, .65
w, h = .7, .3
t = np.linspace(0, chrsize, 1000)
ax = fig.add_axes([xstart, ystart, w, h])
mx, my = zip(*g.scatter_data)
rho = spearmanr(mx, my)
dsg = "g"
ax.vlines(pp, 0, mlgsize, colors="beige")
ax.plot(mx, my, ".", color=set2[3])
ax.plot(t, spl(t), "-", color=dsg)
ax.text(.05, .95, mlg, va="top", transform=ax.transAxes)
normalize_lms_axis(ax,
xlim=chrsize,
ylim=mlgsize,
ylabel="Genetic distance (cM)")
if rho < 0:
ax.invert_yaxis()
# Panel B
ystart -= .28
h = .25
ax = fig.add_axes([xstart, ystart, w, h])
ax.vlines(pp, 0, mlgsize, colors="beige")
ax.plot(t, spld(t), "-", lw=2, color=dsg)
ax.plot(pp, spld(pp), "o", mfc="w", mec=dsg, ms=5)
normalize_lms_axis(ax,
xlim=chrsize,
ylim=25 * 1e-6,
xfactor=1e-6,
xlabel="Physical position (Mb)",
yfactor=1000000,
ylabel="Recomb. rate\n(cM / Mb)")
# Panel C (specific to JMMale-1)
a, b = "scaffold_1076", "scaffold_861"
sizes = dict((x.component_id, (x.object_beg, x.object_end,
x.component_span, x.orientation)) \
for x in g.agp if not x.is_gap)
a_beg, a_end, asize, ao = sizes[a]
b_beg, b_end, bsize, bo = sizes[b]
gapsize = g.get_gapsize(a)
total_size = asize + gapsize + bsize
ratio = .6 / total_size
y = .16
pad = .03
pb_ratio = w / chrsize
# Zoom
lsg = "lightslategray"
root.plot((.15 + pb_ratio * a_beg, .2), (ystart, ystart - .14),
":",
color=lsg)
root.plot((.15 + pb_ratio * b_end, .3), (ystart, ystart - .08),
":",
color=lsg)
ends = []
for tag, size, marker, beg in zip((a, b), (asize, bsize), (49213, 81277),
(.2, .2 + (asize + gapsize) * ratio)):
end = beg + size * ratio
marker = beg + marker * ratio
ends.append((beg, end, marker))
root.plot((marker, ), (y, ), "o", color=lsg)
root.text((beg + end) / 2,
y + pad,
latex(tag),
ha="center",
va="center")
HorizontalChromosome(root, beg, end, y, height=.025, fc='gainsboro')
begs, ends, markers = zip(*ends)
fontprop = dict(color=lsg, ha="center", va="center")
ypos = y + pad * 2
root.plot(markers, (ypos, ypos), "-", lw=2, color=lsg)
root.text(
sum(markers) / 2, ypos + pad,
"Distance: 1.29cM $\Leftrightarrow$ 211,824bp (6.1 cM/Mb)", **fontprop)
ypos = y - pad
xx = markers[0], ends[0]
root.plot(xx, (ypos, ypos), "-", lw=2, color=lsg)
root.text(sum(xx) / 2, ypos - pad, "34,115bp", **fontprop)
xx = markers[1], begs[1]
root.plot(xx, (ypos, ypos), "-", lw=2, color=lsg)
root.text(sum(xx) / 2, ypos - pad, "81,276bp", **fontprop)
root.plot((ends[0], begs[1]), (y, y), ":", lw=2, color=lsg)
root.text(sum(markers) / 2,
ypos - 3 * pad,
r"$\textit{Estimated gap size: 96,433bp}$",
color="r",
ha="center",
va="center")
labels = ((.05, .95, 'A'), (.05, .6, 'B'), (.05, .27, 'C'))
panel_labels(root, labels)
normalize_axes(root)
pf = "estimategaps"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def qc(args):
"""
%prog qc prefix
Expects data files including:
1. `prefix.bedpe` draws Bezier curve between paired reads
2. `prefix.sizes` draws length of the contig/scaffold
3. `prefix.gaps.bed` mark the position of the gaps in sequence
4. `prefix.bed.coverage` plots the base coverage
5. `prefix.pairs.bed.coverage` plots the clone coverage
See assembly.coverage.posmap() for the generation of these files.
"""
from jcvi.graphics.glyph import Bezier
p = OptionParser(qc.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(p.print_help())
prefix, = args
scf = prefix
# All these files *must* be present in the current folder
bedpefile = prefix + ".bedpe"
fastafile = prefix + ".fasta"
sizesfile = prefix + ".sizes"
gapsbedfile = prefix + ".gaps.bed"
bedfile = prefix + ".bed"
bedpefile = prefix + ".bedpe"
pairsbedfile = prefix + ".pairs.bed"
sizes = Sizes(fastafile).mapping
size = sizes[scf]
fig = plt.figure(1, (8, 5))
root = fig.add_axes([0, 0, 1, 1])
# the scaffold
root.add_patch(Rectangle((.1, .15), .8, .03, fc='k'))
# basecoverage and matecoverage
ax = fig.add_axes([.1, .45, .8, .45])
bins = 200 # Smooth the curve
basecoverage = Coverage(bedfile, sizesfile)
matecoverage = Coverage(pairsbedfile, sizesfile)
x, y = basecoverage.get_plot_data(scf, bins=bins)
baseline, = ax.plot(x, y, 'g-')
x, y = matecoverage.get_plot_data(scf, bins=bins)
mateline, = ax.plot(x, y, 'r-')
legends = ("Base coverage", "Mate coverage")
leg = ax.legend((baseline, mateline), legends, shadow=True, fancybox=True)
leg.get_frame().set_alpha(.5)
ax.set_xlim(0, size)
# draw the read pairs
fp = open(bedpefile)
pairs = []
for row in fp:
scf, astart, aend, scf, bstart, bend, clonename = row.split()
astart, bstart = int(astart), int(bstart)
aend, bend = int(aend), int(bend)
start = min(astart, bstart) + 1
end = max(aend, bend)
pairs.append((start, end))
bpratio = .8 / size
cutoff = 1000 # inserts smaller than this are not plotted
# this convert from base => x-coordinate
pos = lambda x: (.1 + x * bpratio)
ypos = .15 + .03
for start, end in pairs:
dist = end - start
if dist < cutoff:
continue
dist = min(dist, 10000)
# 10Kb == .25 canvas height
height = .25 * dist / 10000
xstart = pos(start)
xend = pos(end)
p0 = (xstart, ypos)
p1 = (xstart, ypos + height)
p2 = (xend, ypos + height)
p3 = (xend, ypos)
Bezier(root, p0, p1, p2, p3)
# gaps on the scaffold
fp = open(gapsbedfile)
for row in fp:
b = BedLine(row)
start, end = b.start, b.end
xstart = pos(start)
xend = pos(end)
root.add_patch(Rectangle((xstart, .15), xend - xstart, .03, fc='w'))
root.text(.5, .1, scf, color='b', ha="center")
warn_msg = "Only the inserts > {0}bp are shown".format(cutoff)
root.text(.5, .1, scf, color='b', ha="center")
root.text(.5, .05, warn_msg, color='gray', ha="center")
# clean up and output
set_human_base_axis(ax)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
figname = prefix + ".pdf"
savefig(figname, dpi=300)
def gcdepth(args):
"""
%prog gcdepth sample_name tag
Plot GC content vs depth vs genomnic bins. Inputs are mosdepth output:
- NA12878_S1.mosdepth.global.dist.txt
- NA12878_S1.mosdepth.region.dist.txt
- NA12878_S1.regions.bed.gz
- NA12878_S1.regions.bed.gz.csi
- NA12878_S1.regions.gc.bed.gz
A sample mosdepth.sh script might look like:
```
#!/bin/bash
LD_LIBRARY_PATH=mosdepth/htslib/ mosdepth/mosdepth $1 \\
bams/$1.bam -t 4 -c chr1 -n --by 1000
bedtools nuc -fi GRCh38/WholeGenomeFasta/genome.fa \\
-bed $1.regions.bed.gz \\
| pigz -c > $1.regions.gc.bed.gz
```
"""
import hashlib
from jcvi.algorithms.formula import MAD_interval as confidence_interval
from jcvi.graphics.base import latex, plt, savefig, set2
p = OptionParser(gcdepth.__doc__)
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(not p.print_help())
sample_name, tag = args
# The tag is used to add to title, also provide a random (hashed) color
coloridx = int(hashlib.sha1(tag).hexdigest(), 16) % len(set2)
color = set2[coloridx]
# mosdepth outputs a table that we can use to plot relationship
gcbedgz = sample_name + ".regions.gc.bed.gz"
df = pd.read_csv(gcbedgz, delimiter="\t")
mf = df.loc[:, ("4_usercol", "6_pct_gc")]
mf.columns = ["depth", "gc"]
# We discard any bins that are gaps
mf = mf[(mf["depth"] > .001) | (mf["gc"] > .001)]
# Create GC bins
gcbins = defaultdict(list)
for i, row in mf.iterrows():
gcp = int(round(row["gc"] * 100))
gcbins[gcp].append(row["depth"])
gcd = sorted((k * .01, confidence_interval(v))
for (k, v) in gcbins.items())
gcd_x, gcd_y = zip(*gcd)
m, lo, hi = zip(*gcd_y)
# Plot
plt.plot(mf["gc"], mf["depth"], ".", color="lightslategray", ms=2,
mec="lightslategray", alpha=.1)
patch = plt.fill_between(gcd_x, lo, hi,
facecolor=color, alpha=.25, zorder=10,
linewidth=0.0, label="Median +/- MAD band")
plt.plot(gcd_x, m, "-", color=color, lw=2, zorder=20)
ax = plt.gca()
ax.legend(handles=[patch], loc="best")
ax.set_xlim(0, 1)
ax.set_ylim(0, 100)
ax.set_title("{} ({})".format(latex(sample_name), tag))
ax.set_xlabel("GC content")
ax.set_ylabel("Depth")
savefig(sample_name + ".gcdepth.png")
def demo(args):
"""
%prog demo
Draw sample gene features to illustrate the various fates of duplicate
genes - to be used in a book chapter.
"""
p = OptionParser(demo.__doc__)
opts, args = p.parse_args(args)
fig = plt.figure(1, (8, 5))
root = fig.add_axes([0, 0, 1, 1])
panel_space = 0.23
dup_space = 0.025
# Draw a gene and two regulatory elements at these arbitrary locations
locs = [
(0.5, 0.9), # ancestral gene
(0.5, 0.9 - panel_space + dup_space), # identical copies
(0.5, 0.9 - panel_space - dup_space),
(0.5, 0.9 - 2 * panel_space + dup_space), # degenerate copies
(0.5, 0.9 - 2 * panel_space - dup_space),
(0.2, 0.9 - 3 * panel_space + dup_space), # sub-functionalization
(0.2, 0.9 - 3 * panel_space - dup_space),
(0.5, 0.9 - 3 * panel_space + dup_space), # neo-functionalization
(0.5, 0.9 - 3 * panel_space - dup_space),
(0.8, 0.9 - 3 * panel_space + dup_space), # non-functionalization
(0.8, 0.9 - 3 * panel_space - dup_space),
]
default_regulator = "gm"
regulators = [
default_regulator,
default_regulator,
default_regulator,
"wm",
default_regulator,
"wm",
"gw",
"wb",
default_regulator,
"ww",
default_regulator,
]
width = 0.24
for i, (xx, yy) in enumerate(locs):
regulator = regulators[i]
x1, x2 = xx - 0.5 * width, xx + 0.5 * width
Glyph(root, x1, x2, yy)
if i == 9: # upper copy for non-functionalization
continue
# coding region
x1, x2 = xx - 0.16 * width, xx + 0.45 * width
Glyph(root, x1, x2, yy, fc="k")
# two regulatory elements
x1, x2 = xx - 0.4 * width, xx - 0.28 * width
for xx, fc in zip((x1, x2), regulator):
if fc == "w":
continue
DoubleCircle(root, xx, yy, fc=fc)
rotation = 30
tip = 0.02
if i == 0:
ya = yy + tip
root.text(x1, ya, "Flower", rotation=rotation, va="bottom")
root.text(x2, ya, "Root", rotation=rotation, va="bottom")
elif i == 7:
ya = yy + tip
root.text(x2, ya, "Leaf", rotation=rotation, va="bottom")
# Draw arrows between panels (center)
arrow_dist = 0.08
ar_xpos = 0.5
for ar_ypos in (0.3, 0.53, 0.76):
root.annotate(
" ",
(ar_xpos, ar_ypos),
(ar_xpos, ar_ypos + arrow_dist),
arrowprops=arrowprops,
)
ar_ypos = 0.3
for ar_xpos in (0.2, 0.8):
root.annotate(" ", (ar_xpos, ar_ypos), (0.5, ar_ypos + arrow_dist),
arrowprops=arrowprops)
# Duplication, Degeneration
xx = 0.6
ys = (0.76, 0.53)
processes = ("Duplication", "Degeneration")
for yy, process in zip(ys, processes):
root.text(xx, yy + 0.02, process, fontweight="bold")
# Label of fates
xs = (0.2, 0.5, 0.8)
fates = ("Subfunctionalization", "Neofunctionalization",
"Nonfunctionalization")
yy = 0.05
for xx, fate in zip(xs, fates):
RoundLabel(root, xx, yy, fate)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
figname = "demo.pdf"
savefig(figname, dpi=300)
def plot(args):
"""
%prog plot tagged.new.bed chr1
Plot gene identifiers along a particular chromosome, often to illustrate the
gene id assignment procedure.
"""
from jcvi.graphics.base import plt, savefig
from jcvi.graphics.chromosome import ChromosomeMap
p = OptionParser(plot.__doc__)
p.add_option("--firstn", type="int", help="Only plot the first N genes")
p.add_option("--ymax", type="int", help="Y-axis max value")
p.add_option("--log", action="store_true", help="Write plotting data")
opts, args, iopts = p.set_image_options(args, figsize="6x4")
if len(args) != 2:
sys.exit(not p.print_help())
taggedbed, chr = args
bed = Bed(taggedbed)
beds = list(bed.sub_bed(chr))
old, new = [], []
i = 0
for b in beds:
accn = b.extra[0]
if "te" in accn:
continue
accn, tag = accn.split("|")
if tag == "OVERLAP":
continue
c, r = atg_name(accn)
if tag == "NEW":
new.append((i, r))
else:
old.append((i, r))
i += 1
ngenes = i
assert ngenes == len(new) + len(old)
logging.debug("Imported {0} ranks on {1}.".format(ngenes, chr))
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
xstart, xend = 0.2, 0.8
ystart, yend = 0.2, 0.8
pad = 0.02
ngenes = opts.firstn or ngenes
ymax = opts.ymax or 500000
title = "Assignment of Medtr identifiers"
if opts.ymax:
subtitle = "{0}, first {1} genes".format(chr, ngenes)
else:
subtitle = "{0}, {1} genes ({2} new)".format(chr, ngenes, len(new))
chr_map = ChromosomeMap(fig, root, xstart, xend, ystart, yend, pad, 0,
ymax, 5, title, subtitle)
ax = chr_map.axes
if opts.log:
from jcvi.utils.table import write_csv
header = ["x", "y"]
write_csv(header, new, filename=chr + ".new")
write_csv(header, old, filename=chr + ".old")
x, y = zip(*new)
ax.plot(x, y, "b,")
x, y = zip(*old)
ax.plot(x, y, "r,")
# Legends
ymid = (ystart + yend) / 2
y = ymid + pad
root.plot([0.2], [y], "r.", lw=2)
root.text(0.2 + pad, y, "Existing Medtr ids", va="center", size=10)
y = ymid - pad
root.plot([0.2], [y], "b.", lw=2)
root.text(0.2 + pad, y, "Newly instantiated ids", va="center", size=10)
ax.set_xlim(0, ngenes)
ax.set_ylim(0, ymax)
ax.set_axis_off()
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = chr + ".identifiers." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
ratio = ysize * 1. / xsize if proportional else 1
width = iopts.w
height = iopts.h * ratio
fig = plt.figure(1, (width, height))
root = fig.add_axes([0, 0, 1, 1]) # the whole canvas
ax = fig.add_axes([.1, .1, .8, .8]) # the dot plot
blastplot(ax,
blastfile,
qsizes,
ssizes,
qbed,
sbed,
style=opts.dotstyle,
proportional=proportional,
sampleN=opts.nmax,
baseticks=True,
stripNames=opts.stripNames,
highlights=highlights)
# add genome names
to_ax_label = lambda fname: op.basename(fname).split(".")[0]
gx, gy = [to_ax_label(x.filename) for x in (qsizes, ssizes)]
ax.set_xlabel(gx, size=16)
ax.set_ylabel(gy, size=16)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def main():
p = OptionParser(__doc__)
opts, args, iopts = p.set_image_options(figsize="9x7")
if len(args) != 1:
sys.exit(not p.print_help())
(mode, ) = args
assert mode == "demo"
a, b = 30, 70
pad = 0.08
w = 0.31
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
# Row separators
yy = 1 - pad
for i in range(3):
root.plot((0, 1), (yy, yy), "-", lw=2, color="lightgray")
yy -= w
# Row headers
xx = pad * 0.6
yy = 1 - pad - 0.5 * w
for title in ("Inversion", "Indel", "Duplication"):
root.text(xx, yy, title, ha="center", va="center")
yy -= w
# Column headers
xx = pad + 0.5 * w
yy = 1 - pad / 2
for title in ("Assembly alignment", "Read alignment",
"Optical map alignment"):
root.text(xx, yy, title, ha="center", va="center")
xx += w
p = PairwiseAlign(fig, [pad, 2 * w, w, w])
p.invert(a, b)
p.draw()
p = PairwiseAlign(fig, [pad, w, w, w])
p.delete(a, b)
p.draw()
p = PairwiseAlign(fig, [pad, 0, w, w])
p.duplicate(a, b, gap=5)
p.draw()
p = ReadAlign(fig, [pad + w, 2 * w, w, w])
p.invert(a, b)
p.draw()
p = ReadAlign(fig, [pad + w, w, w, w])
p.delete(a, b)
p.draw()
p = ReadAlign(fig, [pad + w, 0, w, w])
p.duplicate(a, b)
p.draw()
p = OpticalMapAlign(fig, [pad + 2 * w, 2 * w, w, w])
p.invert(a, b)
p.draw()
p = OpticalMapAlign(fig, [pad + 2 * w, w, w, w])
p.delete(a, b)
p.draw()
p = OpticalMapAlign(fig, [pad + 2 * w, 0, w, w])
p.duplicate(a, b)
p.draw()
normalize_axes(root)
image_name = mode + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def cotton(args):
"""
%prog cotton seqids karyotype.layout mcscan.out all.bed synteny.layout
Build a composite figure that calls graphics.karyotype and graphic.synteny.
"""
p = OptionParser(cotton.__doc__)
p.add_option("--depthfile", help="Use depth info in this file")
p.add_option("--switch", help="Rename the seqid with two-column file")
opts, args, iopts = p.set_image_options(args, figsize="8x7")
if len(args) != 5:
sys.exit(p.print_help())
seqidsfile, klayout, datafile, bedfile, slayout = args
switch = opts.switch
depthfile = opts.depthfile
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
kt = Karyotype(fig, root, seqidsfile, klayout)
Synteny(fig, root, datafile, bedfile, slayout, switch=switch)
light = "lightslategrey"
# Show the dup depth along the cotton chromosomes
if depthfile:
ymin, ymax = 0.9, 0.95
root.text(0.11,
0.96,
"Cotton duplication level",
color="gray",
size=10)
root.plot([0.1, 0.95], [ymin, ymin], color="gray")
root.text(0.96, 0.9, "1x", color="gray", va="center")
root.plot([0.1, 0.95], [ymax, ymax], color="gray")
root.text(0.96, 0.95, "6x", color="gray", va="center")
fp = open(depthfile)
track = kt.tracks[0] # Cotton
depths = []
for row in fp:
a, b, depth = row.split()
depth = int(depth)
try:
p = track.get_coords(a)
depths.append((p, depth))
except KeyError:
pass
depths.sort(key=lambda x: (x[0], -x[1]))
xx, yy = zip(*depths)
yy = [ymin + 0.01 * (x - 1) for x in yy]
root.plot(xx, yy, "-", color=light)
# legend showing the orientation of the genes
draw_gene_legend(root, 0.5, 0.68, 0.5)
# Zoom
xpos = 0.835
ytop = 0.9
xmin, xmax = 0.18, 0.82
ymin, ymax = ytop, 0.55
lc = "k"
kwargs = dict(lw=3, color=lc, mec=lc, mfc="w", zorder=3)
root.plot((xpos, xpos), (ymax, 0.63), ":o", **kwargs)
root.plot((xpos, xmin), (ymax, ymin), ":o", **kwargs)
root.plot((xpos, xmax), (ymax, ymin), ":o", **kwargs)
RoundRect(root, (0.06, 0.17), 0.92, 0.35, fill=False, lw=2, ec=light)
# Panels
root.text(0.05, 0.95, "a", size=20, fontweight="bold")
root.text(0.1, 0.45, "b", size=20, fontweight="bold")
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "cotton"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def stack(args):
"""
%prog stack fastafile
Create landscape plots that show the amounts of genic sequences, and repetitive
sequences along the chromosomes.
"""
p = OptionParser(stack.__doc__)
p.add_option("--top", default=10, type="int",
help="Draw the first N chromosomes [default: %default]")
p.add_option("--stacks",
default="Exons,Introns,DNA_transposons,Retrotransposons",
help="Features to plot in stackplot [default: %default]")
p.add_option("--switch",
help="Change chr names based on two-column file [default: %default]")
add_window_options(p)
opts, args, iopts = p.set_image_options(args, figsize="8x8")
if len(args) != 1:
sys.exit(not p.print_help())
fastafile, = args
top = opts.top
window, shift, subtract = check_window_options(opts)
switch = opts.switch
if switch:
switch = DictFile(opts.switch)
stacks = opts.stacks.split(",")
bedfiles = get_beds(stacks)
binfiles = get_binfiles(bedfiles, fastafile, shift, subtract=subtract)
sizes = Sizes(fastafile)
s = list(sizes.iter_sizes())[:top]
maxl = max(x[1] for x in s)
margin = .08
inner = .02 # y distance between tracks
pf = fastafile.rsplit(".", 1)[0]
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
# Gauge
ratio = draw_gauge(root, margin, maxl)
# Per chromosome
yinterval = (1 - 2 * margin) / (top + 1)
xx = margin
yy = 1 - margin
for chr, clen in s:
yy -= yinterval
xlen = clen / ratio
cc = chr
if "_" in chr:
ca, cb = chr.split("_")
cc = ca[0].upper() + cb
if switch and cc in switch:
cc = "\n".join((cc, "({0})".format(switch[cc])))
root.add_patch(Rectangle((xx, yy), xlen, yinterval - inner, color=gray))
ax = fig.add_axes([xx, yy, xlen, yinterval - inner])
nbins = clen / shift
if clen % shift:
nbins += 1
stackplot(ax, binfiles, nbins, palette, chr, window, shift)
root.text(xx - .04, yy + .5 * (yinterval - inner), cc, ha="center", va="center")
ax.set_xlim(0, nbins)
ax.set_ylim(0, 1)
ax.set_axis_off()
# Legends
yy -= yinterval
xx = margin
for b, p in zip(bedfiles, palette):
b = b.rsplit(".", 1)[0].replace("_", " ")
b = Registration.get(b, b)
root.add_patch(Rectangle((xx, yy), inner, inner, color=p, lw=0))
xx += 2 * inner
root.text(xx, yy, b, size=13)
xx += len(b) * .012 + inner
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def waterlilyGOM(args):
"""
%prog mcmctree.tre table.csv
Customized figure to plot phylogeny and related infographics.
"""
from jcvi.graphics.tree import (
LeafInfoFile,
WGDInfoFile,
draw_tree,
parse_tree,
draw_wgd_xy,
)
from jcvi.graphics.table import CsvTable, draw_table
p = OptionParser(waterlilyGOM.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="12x9")
if len(args) != 2:
sys.exit(not p.print_help())
(datafile, csvfile) = args
outgroup = ["ginkgo"]
logging.debug("Load tree file `{0}`".format(datafile))
t, hpd = parse_tree(datafile)
pf = datafile.rsplit(".", 1)[0]
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
margin, rmargin = 0.15, 0.19 # Left and right margin
leafinfo = LeafInfoFile("leafinfo.csv").cache
wgdinfo = WGDInfoFile("wgdinfo.csv").cache
groups = "Monocots,Eudicots,ANA-grade,Gymnosperms"
draw_tree(
root,
t,
hpd=hpd,
margin=margin,
rmargin=rmargin,
supportcolor=None,
internal=False,
outgroup=outgroup,
leafinfo=leafinfo,
wgdinfo=wgdinfo,
geoscale=True,
groups=groups.split(","),
)
# Bottom right show legends for the WGD circles
pad = 0.02
ypad = 0.04
xstart = 1 - rmargin + pad
ystart = 0.2
waterlily_wgdline = wgdinfo["waterlily"][0]
ypos = ystart - 2 * ypad
draw_wgd_xy(root, xstart, ypos, waterlily_wgdline)
root.text(
xstart + pad,
ypos,
"Nymphaealean WGD",
color=waterlily_wgdline.color,
va="center",
)
other_wgdline = wgdinfo["banana"][0]
ypos = ystart - 3 * ypad
draw_wgd_xy(root, xstart, ypos, other_wgdline)
root.text(
xstart + pad,
ypos,
"Other known WGDs",
color=other_wgdline.color,
va="center",
)
# Top left draw the comparison table
csv_table = CsvTable(csvfile)
draw_table(
root,
csv_table,
extent=(0.02, 0.44, 0.55, 0.985),
stripe_color="lavender",
yinflation=iopts.w / iopts.h,
)
normalize_axes(root)
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def composite(args):
"""
%prog composite fastafile chr1
Combine line plots, feature bars and alt-bars, different data types
specified in options. Inputs must be BED-formatted. Three types of viz are
currently supported:
--lines: traditional line plots, useful for plotting feature freq
--bars: show where the extent of features are
--altbars: similar to bars, yet in two alternating tracks, e.g. scaffolds
"""
from jcvi.graphics.chromosome import HorizontalChromosome
p = OptionParser(composite.__doc__)
p.add_option("--lines",
help="Features to plot in lineplot [default: %default]")
p.add_option("--bars",
help="Features to plot in bars [default: %default]")
p.add_option("--altbars",
help="Features to plot in alt-bars [default: %default]")
p.add_option("--fatten", default=False, action="store_true",
help="Help visualize certain narrow features [default: %default]")
p.add_option("--mode", default="span", choices=("span", "count", "score"),
help="Accumulate feature based on [default: %default]")
add_window_options(p)
opts, args, iopts = p.set_image_options(args, figsize="8x5")
if len(args) != 2:
sys.exit(not p.print_help())
fastafile, chr = args
window, shift, subtract = check_window_options(opts)
linebeds, barbeds, altbarbeds = [], [], []
fatten = opts.fatten
if opts.lines:
lines = opts.lines.split(",")
linebeds = get_beds(lines)
if opts.bars:
bars = opts.bars.split(",")
barbeds = get_beds(bars)
if opts.altbars:
altbars = opts.altbars.split(",")
altbarbeds = get_beds(altbars)
linebins = get_binfiles(linebeds, fastafile, shift, mode=opts.mode)
margin = .12
clen = Sizes(fastafile).mapping[chr]
nbins = get_nbins(clen, shift)
plt.rcParams["xtick.major.size"] = 0
plt.rcParams["ytick.major.size"] = 0
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
root.text(.5, .95, chr, ha="center", color="darkslategray")
xstart, xend = margin, 1 - margin
xlen = xend - xstart
ratio = xlen / clen
# Line plots
ax = fig.add_axes([xstart, .6, xlen, .3])
lineplot(ax, linebins, nbins, chr, window, shift)
# Bar plots
yy = .5
yinterval = .08
xs = lambda x: xstart + ratio * x
r = .01
fattend = .0025
for bb in barbeds:
root.text(xend + .01, yy, bb.split(".")[0], va="center")
HorizontalChromosome(root, xstart, xend, yy, height=.02)
bb = Bed(bb)
for b in bb:
start, end = xs(b.start), xs(b.end)
span = end - start
if fatten and span < fattend:
span = fattend
root.add_patch(Rectangle((start, yy - r), span, 2 * r, \
lw=0, fc="darkslategray"))
yy -= yinterval
# Alternative bar plots
offset = r / 2
for bb in altbarbeds:
root.text(xend + .01, yy, bb.split(".")[0], va="center")
bb = Bed(bb)
for i, b in enumerate(bb):
start, end = xs(b.start), xs(b.end)
span = end - start
if span < .0001:
continue
offset = -offset
root.add_patch(Rectangle((start, yy + offset), end - start, .003, \
lw=0, fc="darkslategray"))
yy -= yinterval
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = chr + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def oropetium(args):
"""
%prog oropetium mcscan.out all.bed layout switch.ids
Build a composite figure that calls graphis.synteny.
"""
p = OptionParser(oropetium.__doc__)
p.add_option("--extra", help="Extra features in BED format")
opts, args, iopts = p.set_image_options(args, figsize="9x6")
if len(args) != 4:
sys.exit(not p.print_help())
datafile, bedfile, slayout, switch = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
Synteny(fig,
root,
datafile,
bedfile,
slayout,
switch=switch,
extra_features=opts.extra)
# legend showing the orientation of the genes
draw_gene_legend(root, 0.4, 0.57, 0.74, text=True, repeat=True)
# On the left panel, make a species tree
fc = "lightslategrey"
coords = {}
xs, xp = 0.16, 0.03
coords["oropetium"] = (xs, 0.7)
coords["setaria"] = (xs, 0.6)
coords["sorghum"] = (xs, 0.5)
coords["rice"] = (xs, 0.4)
coords["brachypodium"] = (xs, 0.3)
xs -= xp
coords["Panicoideae"] = join_nodes(root, coords, "setaria", "sorghum", xs)
xs -= xp
coords["BEP"] = join_nodes(root, coords, "rice", "brachypodium", xs)
coords["PACMAD"] = join_nodes(root, coords, "oropetium", "Panicoideae", xs)
xs -= xp
coords["Poaceae"] = join_nodes(root, coords, "BEP", "PACMAD", xs)
# Names of the internal nodes
for tag in ("BEP", "Poaceae"):
nx, ny = coords[tag]
nx, ny = nx - 0.005, ny - 0.02
root.text(nx, ny, tag, rotation=90, ha="right", va="top", color=fc)
for tag in ("PACMAD", ):
nx, ny = coords[tag]
nx, ny = nx - 0.005, ny + 0.02
root.text(nx, ny, tag, rotation=90, ha="right", va="bottom", color=fc)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "oropetium"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def bites(args):
"""
%prog bites
Illustrate the pipeline for automated bite discovery.
"""
p = OptionParser(__doc__)
opts, args = p.parse_args()
fig = plt.figure(1, (6, 6))
root = fig.add_axes([0, 0, 1, 1])
# HSP pairs
hsps = (
((50, 150), (60, 180)),
((190, 250), (160, 235)),
((300, 360), (270, 330)),
((430, 470), (450, 490)),
((570, 620), (493, 543)),
((540, 555), (370, 385)), # non-collinear hsps
)
titlepos = (0.9, 0.65, 0.4)
titles = ("Compare orthologous region", "Find collinear HSPs",
"Scan paired gaps")
ytip = 0.01
mrange = 650.0
m = lambda x: x / mrange * 0.7 + 0.1
for i, (ya, title) in enumerate(zip(titlepos, titles)):
yb = ya - 0.1
plt.plot((0.1, 0.8), (ya, ya), "-", color="gray", lw=2, zorder=1)
plt.plot((0.1, 0.8), (yb, yb), "-", color="gray", lw=2, zorder=1)
RoundLabel(root, 0.5, ya + 4 * ytip, title)
root.text(0.9, ya, "A. thaliana", ha="center", va="center")
root.text(0.9, yb, "B. rapa", ha="center", va="center")
myhsps = hsps
if i >= 1:
myhsps = hsps[:-1]
for (a, b), (c, d) in myhsps:
a, b, c, d = [m(x) for x in (a, b, c, d)]
r1 = Rectangle((a, ya - ytip),
b - a,
2 * ytip,
fc="r",
lw=0,
zorder=2)
r2 = Rectangle((c, yb - ytip),
d - c,
2 * ytip,
fc="r",
lw=0,
zorder=2)
r3 = Rectangle((a, ya - ytip),
b - a,
2 * ytip,
fill=False,
zorder=3)
r4 = Rectangle((c, yb - ytip),
d - c,
2 * ytip,
fill=False,
zorder=3)
r5 = Polygon(
((a, ya - ytip), (c, yb + ytip), (d, yb + ytip),
(b, ya - ytip)),
fc="r",
alpha=0.2,
)
rr = (r1, r2, r3, r4, r5)
if i == 2:
rr = rr[:-1]
for r in rr:
root.add_patch(r)
# Gap pairs
hspa, hspb = zip(*myhsps)
gapa, gapb = [], []
for (a, b), (c, d) in pairwise(hspa):
gapa.append((b + 1, c - 1))
for (a, b), (c, d) in pairwise(hspb):
gapb.append((b + 1, c - 1))
gaps = zip(gapa, gapb)
tpos = titlepos[-1]
yy = tpos - 0.05
for i, ((a, b), (c, d)) in enumerate(gaps):
i += 1
a, b, c, d = [m(x) for x in (a, b, c, d)]
xx = (a + b + c + d) / 4
TextCircle(root, xx, yy, str(i))
# Bites
ystart = 0.24
ytip = 0.05
bites = (
("Bite(40=>-15)", True),
("Bite(50=>35)", False),
("Bite(70=>120)", False),
("Bite(100=>3)", True),
)
for i, (bite, selected) in enumerate(bites):
xx = 0.15 if (i % 2 == 0) else 0.55
yy = ystart - i / 2 * ytip
i += 1
TextCircle(root, xx, yy, str(i))
color = "k" if selected else "gray"
root.text(xx + ytip, yy, bite, size=10, color=color, va="center")
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
figname = fname() + ".pdf"
savefig(figname, dpi=300)
def birch(args):
"""
%prog birch seqids layout
Plot birch macro-synteny, with an embedded phylogenetic tree to the right.
"""
p = OptionParser(birch.__doc__)
opts, args, iopts = p.set_image_options(args, figsize="8x6")
if len(args) != 2:
sys.exit(not p.print_help())
seqids, layout = args
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
K = Karyotype(fig, root, seqids, layout)
L = K.layout
xs = 0.79
dt = dict(rectangle=False, circle=False)
# Embed a phylogenetic tree to the right
coords = {}
coords["Amborella"] = (xs, L[0].y)
coords["Vitis"] = (xs, L[1].y)
coords["Prunus"] = (xs, L[2].y)
coords["Betula"] = (xs, L[3].y)
coords["Populus"] = (xs, L[4].y)
coords["Arabidopsis"] = (xs, L[5].y)
coords["fabids"] = join_nodes(root, coords, "Prunus", "Betula", xs, **dt)
coords["malvids"] = join_nodes(root, coords, "Populus", "Arabidopsis", xs,
**dt)
coords["rosids"] = join_nodes(root, coords, "fabids", "malvids", xs, **dt)
coords["eudicots"] = join_nodes(root, coords, "rosids", "Vitis", xs, **dt)
coords["angiosperm"] = join_nodes(root, coords, "eudicots", "Amborella",
xs, **dt)
# Show branch length
branch_length(root, coords["Amborella"], coords["angiosperm"], ">160.0")
branch_length(root,
coords["eudicots"],
coords["angiosperm"],
">78.2",
va="top")
branch_length(root, coords["Vitis"], coords["eudicots"], "138.5")
branch_length(root, coords["rosids"], coords["eudicots"], "19.8", va="top")
branch_length(root,
coords["Prunus"],
coords["fabids"],
"104.2",
ha="right",
va="top")
branch_length(root,
coords["Arabidopsis"],
coords["malvids"],
"110.2",
va="top")
branch_length(root,
coords["fabids"],
coords["rosids"],
"19.8",
ha="right",
va="top")
branch_length(root, coords["malvids"], coords["rosids"], "8.5", va="top")
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
pf = "birch"
image_name = pf + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def depth(args):
"""
%prog depth anchorfile --qbed qbedfile --sbed sbedfile
Calculate the depths in the two genomes in comparison, given in --qbed and
--sbed. The synteny blocks will be layered on the genomes, and the
multiplicity will be summarized to stderr.
"""
from jcvi.utils.range import range_depth
p = OptionParser(depth.__doc__)
p.add_option("--depthfile",
help="Generate file with gene and depth [default: %default]")
p.add_option("--histogram", default=False, action="store_true",
help="Plot histograms in PDF")
p.add_option("--xmax", type="int", help="x-axis maximum to display in plot")
p.add_option("--title", default=None, help="Title to display in plot")
p.add_option("--quota", help="Force to use this quota, e.g. 1:1, 1:2 ...")
p.set_beds()
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
anchorfile, = args
qbed, sbed, qorder, sorder, is_self = check_beds(anchorfile, p, opts)
depthfile = opts.depthfile
ac = AnchorFile(anchorfile)
qranges = []
sranges = []
blocks = ac.blocks
for ib in blocks:
q, s, t = zip(*ib)
q = [qorder[x] for x in q]
s = [sorder[x] for x in s]
qrange = (min(q)[0], max(q)[0])
srange = (min(s)[0], max(s)[0])
qranges.append(qrange)
sranges.append(srange)
if is_self:
qranges.append(srange)
qgenome = op.basename(qbed.filename).split(".")[0]
sgenome = op.basename(sbed.filename).split(".")[0]
qtag = "Genome {0} depths".format(qgenome)
print("{}:".format(qtag), file=sys.stderr)
dsq, details = range_depth(qranges, len(qbed))
if depthfile:
fw = open(depthfile, "w")
write_details(fw, details, qbed)
if is_self:
return
stag = "Genome {0} depths".format(sgenome)
print("{}:".format(stag), file=sys.stderr)
dss, details = range_depth(sranges, len(sbed))
if depthfile:
write_details(fw, details, sbed)
fw.close()
logging.debug("Depth written to `{0}`.".format(depthfile))
if not opts.histogram:
return
from jcvi.graphics.base import plt, quickplot_ax, savefig, normalize_axes
# Plot two histograms one for query genome, one for subject genome
plt.figure(1, (6, 3))
f, (ax1, ax2) = plt.subplots(1, 2, sharey=True)
xmax = opts.xmax or max(4, max(list(dsq.keys()) + list(dss.keys())))
if opts.quota:
speak, qpeak = opts.quota.split(":")
qpeak, speak = int(qpeak), int(speak)
else:
qpeak = find_peak(dsq)
speak = find_peak(dss)
qtag = "# of {} blocks per {} gene".format(sgenome, qgenome)
stag = "# of {} blocks per {} gene".format(qgenome, sgenome)
quickplot_ax(ax1, dss, 0, xmax, stag, ylabel="Percentage of genome",
highlight=range(1, speak + 1))
quickplot_ax(ax2, dsq, 0, xmax, qtag, ylabel=None,
highlight=range(1, qpeak + 1))
title = opts.title or "{} vs {} syntenic depths\n{}:{} pattern"\
.format(qgenome, sgenome, speak, qpeak)
root = f.add_axes([0, 0, 1, 1])
vs, pattern = title.split('\n')
root.text(.5, .97, vs, ha="center", va="center", color="darkslategray")
root.text(.5, .925, pattern, ha="center", va="center",
color="tomato", size=16)
print(title, file=sys.stderr)
normalize_axes(root)
pf = anchorfile.rsplit(".", 1)[0] + ".depth"
image_name = pf + ".pdf"
savefig(image_name)
def seeds(args):
"""
%prog seeds [pngfile|jpgfile]
Extract seed metrics from [pngfile|jpgfile]. Use --rows and --cols to crop image.
"""
p = OptionParser(seeds.__doc__)
p.set_outfile()
opts, args, iopts = add_seeds_options(p, args)
if len(args) != 1:
sys.exit(not p.print_help())
(pngfile,) = args
pf = opts.prefix or op.basename(pngfile).rsplit(".", 1)[0]
sigma, kernel = opts.sigma, opts.kernel
rows, cols = opts.rows, opts.cols
labelrows, labelcols = opts.labelrows, opts.labelcols
ff = opts.filter
calib = opts.calibrate
outdir = opts.outdir
if outdir != ".":
mkdir(outdir)
if calib:
calib = json.load(must_open(calib))
pixel_cm_ratio, tr = calib["PixelCMratio"], calib["RGBtransform"]
tr = np.array(tr)
nbcolor = opts.changeBackground
pngfile = convert_background(pngfile, nbcolor)
resizefile, mainfile, labelfile, exif = convert_image(
pngfile,
pf,
outdir=outdir,
rotate=opts.rotate,
rows=rows,
cols=cols,
labelrows=labelrows,
labelcols=labelcols,
)
oimg = load_image(resizefile)
img = load_image(mainfile)
fig, (ax1, ax2, ax3, ax4) = plt.subplots(
ncols=4, nrows=1, figsize=(iopts.w, iopts.h)
)
# Edge detection
img_gray = rgb2gray(img)
logging.debug("Running {0} edge detection ...".format(ff))
if ff == "canny":
edges = canny(img_gray, sigma=opts.sigma)
elif ff == "roberts":
edges = roberts(img_gray)
elif ff == "sobel":
edges = sobel(img_gray)
edges = clear_border(edges, buffer_size=opts.border)
selem = disk(kernel)
closed = closing(edges, selem) if kernel else edges
filled = binary_fill_holes(closed)
# Watershed algorithm
if opts.watershed:
distance = distance_transform_edt(filled)
local_maxi = peak_local_max(distance, threshold_rel=0.05, indices=False)
coordinates = peak_local_max(distance, threshold_rel=0.05)
markers, nmarkers = label(local_maxi, return_num=True)
logging.debug("Identified {0} watershed markers".format(nmarkers))
labels = watershed(closed, markers, mask=filled)
else:
labels = label(filled)
# Object size filtering
w, h = img_gray.shape
canvas_size = w * h
min_size = int(round(canvas_size * opts.minsize / 100))
max_size = int(round(canvas_size * opts.maxsize / 100))
logging.debug(
"Find objects with pixels between {0} ({1}%) and {2} ({3}%)".format(
min_size, opts.minsize, max_size, opts.maxsize
)
)
# Plotting
ax1.set_title("Original picture")
ax1.imshow(oimg)
params = "{0}, $\sigma$={1}, $k$={2}".format(ff, sigma, kernel)
if opts.watershed:
params += ", watershed"
ax2.set_title("Edge detection\n({0})".format(params))
closed = gray2rgb(closed)
ax2_img = labels
if opts.edges:
ax2_img = closed
elif opts.watershed:
ax2.plot(coordinates[:, 1], coordinates[:, 0], "g.")
ax2.imshow(ax2_img, cmap=iopts.cmap)
ax3.set_title("Object detection")
ax3.imshow(img)
filename = op.basename(pngfile)
if labelfile:
accession = extract_label(labelfile)
else:
accession = pf
# Calculate region properties
rp = regionprops(labels)
rp = [x for x in rp if min_size <= x.area <= max_size]
nb_labels = len(rp)
logging.debug("A total of {0} objects identified.".format(nb_labels))
objects = []
for i, props in enumerate(rp):
i += 1
if i > opts.count:
break
y0, x0 = props.centroid
orientation = props.orientation
major, minor = props.major_axis_length, props.minor_axis_length
major_dx = cos(orientation) * major / 2
major_dy = sin(orientation) * major / 2
minor_dx = sin(orientation) * minor / 2
minor_dy = cos(orientation) * minor / 2
ax2.plot((x0 - major_dx, x0 + major_dx), (y0 + major_dy, y0 - major_dy), "r-")
ax2.plot((x0 - minor_dx, x0 + minor_dx), (y0 - minor_dy, y0 + minor_dy), "r-")
npixels = int(props.area)
# Sample the center of the blob for color
d = min(int(round(minor / 2 * 0.35)) + 1, 50)
x0d, y0d = int(round(x0)), int(round(y0))
square = img[(y0d - d) : (y0d + d), (x0d - d) : (x0d + d)]
pixels = []
for row in square:
pixels.extend(row)
logging.debug(
"Seed #{0}: {1} pixels ({2} sampled) - {3:.2f}%".format(
i, npixels, len(pixels), 100.0 * npixels / canvas_size
)
)
rgb = pixel_stats(pixels)
objects.append(Seed(filename, accession, i, rgb, props, exif))
minr, minc, maxr, maxc = props.bbox
rect = Rectangle(
(minc, minr), maxc - minc, maxr - minr, fill=False, ec="w", lw=1
)
ax3.add_patch(rect)
mc, mr = (minc + maxc) / 2, (minr + maxr) / 2
ax3.text(mc, mr, "{0}".format(i), color="w", ha="center", va="center", size=6)
for ax in (ax2, ax3):
ax.set_xlim(0, h)
ax.set_ylim(w, 0)
# Output identified seed stats
ax4.text(0.1, 0.92, "File: {0}".format(latex(filename)), color="g")
ax4.text(0.1, 0.86, "Label: {0}".format(latex(accession)), color="m")
yy = 0.8
fw = must_open(opts.outfile, "w")
if not opts.noheader:
print(Seed.header(calibrate=calib), file=fw)
for o in objects:
if calib:
o.calibrate(pixel_cm_ratio, tr)
print(o, file=fw)
i = o.seedno
if i > 7:
continue
ax4.text(0.01, yy, str(i), va="center", bbox=dict(fc="none", ec="k"))
ax4.text(0.1, yy, o.pixeltag, va="center")
yy -= 0.04
ax4.add_patch(
Rectangle((0.1, yy - 0.025), 0.12, 0.05, lw=0, fc=rgb_to_hex(o.rgb))
)
ax4.text(0.27, yy, o.hashtag, va="center")
yy -= 0.06
ax4.text(
0.1,
yy,
"(A total of {0} objects displayed)".format(nb_labels),
color="darkslategray",
)
normalize_axes(ax4)
for ax in (ax1, ax2, ax3):
xticklabels = [int(x) for x in ax.get_xticks()]
yticklabels = [int(x) for x in ax.get_yticks()]
ax.set_xticklabels(xticklabels, family="Helvetica", size=8)
ax.set_yticklabels(yticklabels, family="Helvetica", size=8)
image_name = op.join(outdir, pf + "." + iopts.format)
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
return objects