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, 0.42, 0.52, 0.48)
labels = ((0.04, 0.96, "A"), (0.04, 0.52, "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 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 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 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 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 [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(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 = .9, .95
root.text(.11, .96, "Cotton duplication level", color="gray", size=10)
root.plot([.1, .95], [ymin, ymin], color="gray")
root.text(.96, .9, "1x", color="gray", va="center")
root.plot([.1, .95], [ymax, ymax], color="gray")
root.text(.96, .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 + .01 * (x - 1) for x in yy]
root.plot(xx, yy, "-", color=light)
# legend showing the orientation of the genes
draw_gene_legend(root, .5, .68, .5)
# Zoom
xpos = .835
ytop = .9
xmin, xmax = .18, .82
ymin, ymax = ytop, .55
lc = "k"
kwargs = dict(lw=3, color=lc, mec=lc, mfc="w", zorder=3)
root.plot((xpos, xpos), (ymax, .63), ":o", **kwargs)
root.plot((xpos, xmin), (ymax, ymin), ":o", **kwargs)
root.plot((xpos, xmax), (ymax, ymin), ":o", **kwargs)
RoundRect(root, (.06, .17), .92, .35, fill=False, lw=2, ec=light)
# Panels
root.text(.05, .95, "a", size=20, fontweight="bold")
root.text(.1, .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 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 fig3(args):
"""
%prog fig3 chrA02,A02,C2,chrC02 chr.sizes all.bed data
Napus Figure 3 displays alignments between quartet chromosomes, inset
with read histograms.
"""
from jcvi.formats.bed import Bed
p = OptionParser(fig3.__doc__)
p.add_option("--gauge_step",
default=10000000,
type="int",
help="Step size for the base scale")
opts, args, iopts = p.set_image_options(args, figsize="12x9")
if len(args) != 4:
sys.exit(not p.print_help())
chrs, sizes, bedfile, datadir = args
gauge_step = opts.gauge_step
diverge = iopts.diverge
rr, gg = diverge
chrs = [[x] for x in chrs.split(",")]
sizes = Sizes(sizes).mapping
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
chr_sizes, chr_sum_sizes, ratio = calc_ratio(chrs, sizes)
# Synteny panel
seqidsfile = make_seqids(chrs)
klayout = make_layout(chrs, chr_sum_sizes, ratio, template_f3a, shift=.05)
height = .07
r = height / 4
K = Karyotype(fig,
root,
seqidsfile,
klayout,
gap=gap,
height=height,
lw=2,
generank=False,
sizes=sizes,
heightpad=r,
roundrect=True,
plot_label=False)
# Chromosome labels
for kl in K.layout:
if kl.empty:
continue
lx, ly = kl.xstart, kl.y
if lx < .11:
lx += .1
ly += .06
label = kl.label
root.text(lx - .015, ly, label, fontsize=15, ha="right", va="center")
# Inset with datafiles
datafiles = ("chrA02.bzh.forxmgr", "parent.A02.per10kb.forxmgr",
"parent.C2.per10kb.forxmgr", "chrC02.bzh.forxmgr")
datafiles = [op.join(datadir, x) for x in datafiles]
tracks = K.tracks
hlfile = op.join(datadir, "bzh.regions.forhaibao")
xy_axes = []
for t, datafile in zip(tracks, datafiles):
ax = make_affix_axis(fig, t, -r, height=2 * r)
xy_axes.append(ax)
chr = t.seqids[0]
xy = XYtrack(ax, datafile, color="lightslategray")
start, end = 0, t.total
xy.interpolate(end)
xy.cap(ymax=40)
xy.import_hlfile(hlfile, chr, diverge=diverge)
xy.draw()
ax.set_xlim(start, end)
gauge_ax = make_affix_axis(fig, t, -r)
adjust_spines(gauge_ax, ["bottom"])
setup_gauge_ax(gauge_ax, start, end, gauge_step)
# Converted gene tracks
ax_Ar = make_affix_axis(fig, tracks[1], r, height=r / 2)
ax_Co = make_affix_axis(fig, tracks[2], r, height=r / 2)
order = Bed(bedfile).order
for asterisk in (False, True):
conversion_track(order,
"data/Genes.Converted.seuil.0.6.AtoC.txt",
0,
"A02",
ax_Ar,
rr,
asterisk=asterisk)
conversion_track(order,
"data/Genes.Converted.seuil.0.6.AtoC.txt",
1,
"C2",
ax_Co,
gg,
asterisk=asterisk)
conversion_track(order,
"data/Genes.Converted.seuil.0.6.CtoA.txt",
0,
"A02",
ax_Ar,
gg,
ypos=1,
asterisk=asterisk)
conversion_track(order,
"data/Genes.Converted.seuil.0.6.CtoA.txt",
1,
"C2",
ax_Co,
rr,
ypos=1,
asterisk=asterisk)
Ar, Co = xy_axes[1:3]
annotations = ((Ar, "Bra028920 Bra028897", "center",
"1DAn2+"), (Ar, "Bra020081 Bra020171", "right", "2DAn2+"),
(Ar, "Bra020218 Bra020286", "left",
"3DAn2+"), (Ar, "Bra008143 Bra008167", "left", "4DAn2-"),
(Ar, "Bra029317 Bra029251", "right",
"5DAn2+ (GSL)"), (Co, "Bo2g001000 Bo2g001300", "left",
"1DCn2-"), (Co, "Bo2g018560 Bo2g023700",
"right", "2DCn2-"),
(Co, "Bo2g024450 Bo2g025390", "left",
"3DCn2-"), (Co, "Bo2g081060 Bo2g082340", "left", "4DCn2+"),
(Co, "Bo2g161510 Bo2g164260", "right", "5DCn2-"))
for ax, genes, ha, label in annotations:
g1, g2 = genes.split()
x1, x2 = order[g1][1].start, order[g2][1].start
if ha == "center":
x = (x1 + x2) / 2 * .8
elif ha == "left":
x = x2
else:
x = x1
label = r"\textit{{{0}}}".format(label)
color = rr if "+" in label else gg
ax.text(x, 30, label, color=color, fontsize=9, ha=ha, va="center")
ax_Ar.set_xlim(0, tracks[1].total)
ax_Ar.set_ylim(-1, 1)
ax_Co.set_xlim(0, tracks[2].total)
ax_Co.set_ylim(-1, 1)
# Plot coverage in resequencing lines
gstep = 5000000
order = "swede,kale,h165,yudal,aviso,abu,bristol".split(",")
labels_dict = {"h165": "Resynthesized (H165)", "abu": "Aburamasari"}
hlsuffix = "regions.forhaibao"
chr1, chr2 = "chrA02", "chrC02"
t1, t2 = tracks[0], tracks[-1]
s1, s2 = sizes[chr1], sizes[chr2]
canvas1 = (t1.xstart, .75, t1.xend - t1.xstart, .2)
c = Coverage(fig,
root,
canvas1,
chr1, (0, s1),
datadir,
order=order,
gauge=None,
plot_chr_label=False,
gauge_step=gstep,
palette="gray",
cap=40,
hlsuffix=hlsuffix,
labels_dict=labels_dict,
diverge=diverge)
yys = c.yys
x1, x2 = .37, .72
tip = .02
annotations = ((x1, yys[2] + .3 * tip, tip, tip / 2,
"FLC"), (x1, yys[3] + .6 * tip, tip, tip / 2, "FLC"),
(x1, yys[5] + .6 * tip, tip, tip / 2,
"FLC"), (x2, yys[0] + .9 * tip, -1.2 * tip, 0, "GSL"),
(x2, yys[4] + .9 * tip, -1.2 * tip, 0,
"GSL"), (x2, yys[6] + .9 * tip, -1.2 * tip, 0, "GSL"))
arrowprops = dict(facecolor='black',
shrink=.05,
frac=.5,
width=1,
headwidth=4)
for x, y, dx, dy, label in annotations:
label = r"\textit{{{0}}}".format(label)
root.annotate(label,
xy=(x, y),
xytext=(x + dx, y + dy),
arrowprops=arrowprops,
color=rr,
fontsize=9,
ha="center",
va="center")
canvas2 = (t2.xstart, .05, t2.xend - t2.xstart, .2)
Coverage(fig,
root,
canvas2,
chr2, (0, s2),
datadir,
order=order,
gauge=None,
plot_chr_label=False,
gauge_step=gstep,
palette="gray",
cap=40,
hlsuffix=hlsuffix,
labels_dict=labels_dict,
diverge=diverge)
pad = .03
labels = ((.1, .67, "A"), (t1.xstart - 3 * pad, .95 + pad, "B"),
(t2.xstart - 3 * pad, .25 + pad, "C"))
panel_labels(root, labels)
normalize_axes(root)
image_name = "napus-fig3." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def cov(args):
"""
%prog cov chrA01 chrC01 chr.sizes data AN.CN.1x1.lifted.anchors.simple
Plot coverage graphs between homeologs, the middle panel show the
homeologous gene pairs. Allow multiple chromosomes to multiple chromosomes.
"""
p = OptionParser(cov.__doc__)
p.add_option("--order",
default="swede,kale,h165,yudal,aviso,abu,bristol,bzh",
help="The order to plot the tracks, comma-separated")
p.add_option("--reverse",
default=False,
action="store_true",
help="Plot the order in reverse")
p.add_option("--gauge_step",
default=5000000,
type="int",
help="Step size for the base scale")
p.add_option(
"--hlsuffix",
default="regions.forhaibao",
help="Suffix for the filename to be used to highlight regions")
opts, args, iopts = p.set_image_options(args, figsize="11x8")
if len(args) != 4:
sys.exit(not p.print_help())
chr1, chr2, sizesfile, datadir = args
chr1 = chr1.split(",")
chr2 = chr2.split(",")
order = opts.order
hlsuffix = opts.hlsuffix
if order:
order = order.split(",")
if opts.reverse:
order.reverse()
sizes = Sizes(sizesfile).mapping
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
chrs = (chr1, chr2)
chr_sizes, chr_sum_sizes, ratio = calc_ratio(chrs, sizes)
chr_size1, chr_size2 = chr_sum_sizes
chr_sizes1, chr_sizes2 = chr_sizes
w1_start, w1_end = center_panel(chr1, chr_size1, ratio)
w2_start, w2_end = center_panel(chr2, chr_size2, ratio)
w1s = w1_start
w2s = w2_start
dsg = "gray"
i = 0
for c1, s1 in zip(chr1, chr_sizes1):
w1 = ratio * s1
plot_label = i == 0
i += 1
canvas1 = (w1s, .6, w1, .3)
Coverage(fig,
root,
canvas1,
c1, (0, s1),
datadir,
order=order,
gauge="top",
plot_label=plot_label,
gauge_step=opts.gauge_step,
palette=dsg,
cap=40,
hlsuffix=hlsuffix)
w1s += w1 + gap
i = 0
for c2, s2 in zip(chr2, chr_sizes2):
w2 = ratio * s2
plot_label = i == 0
i += 1
canvas2 = (w2s, .15, w2, .3)
Coverage(fig,
root,
canvas2,
c2, (0, s2),
datadir,
order=order,
gauge="bottom",
plot_label=plot_label,
gauge_step=opts.gauge_step,
palette=dsg,
cap=40,
hlsuffix=hlsuffix)
w2s += w2 + gap
# Synteny panel
seqidsfile = make_seqids(chrs)
klayout = make_layout(chrs, chr_sum_sizes, ratio, template_cov)
Karyotype(fig,
root,
seqidsfile,
klayout,
gap=gap,
generank=False,
sizes=sizes)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
chr2 = "_".join(chr2)
if opts.reverse:
chr2 += ".reverse"
image_name = chr2 + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)