def main():
p = OptionParser(__doc__)
p.add_option(
"--customfont",
default="Airswing.ttf",
choices=available_fonts,
help="Custom font name",
)
p.add_option("--color", default="limegreen", help="Font color")
p.add_option("--size", default=36, type="int", help="Font size")
opts, args, iopts = p.set_image_options(figsize="2x1", dpi=60, format="png")
if len(args) != 1:
sys.exit(not p.print_help())
(text,) = args
plt.rcdefaults()
fig = plt.figure(1, (iopts.w, iopts.h))
ax = fig.add_axes([0, 0, 1, 1])
ax.text(0.5, 0.5, text, color=opts.color, ha="center", va="center")
fontprop(ax, opts.customfont, size=opts.size)
ax.set_xlim(0, 1)
ax.set_ylim(0, 1)
ax.set_axis_off()
image_name = text + "." + iopts.format
savefig(image_name, dpi=iopts.dpi, iopts=iopts)
def main(tx=None):
"""
%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.
"""
p = OptionParser(main.__doc__)
p.add_option("--outgroup", help="Root the tree using the outgroup. " + \
"Use comma to separate multiple taxa.")
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]")
opts, args, iopts = set_image_options(p, figsize="8x6")
if len(args) != 1:
sys.exit(not p.print_help())
datafile, = args
outgroup = None
if opts.outgroup:
outgroup = opts.outgroup.split(",")
pf = datafile.rsplit(".", 1)[0]
if tx:
pf = "demo"
else:
tx = open(datafile).read()
logging.debug("Load tree file `{0}`.".format(datafile))
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
draw_tree(root,
tx,
rmargin=opts.rmargin,
outgroup=outgroup,
gffdir=opts.gffdir,
sizes=opts.sizes)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = pf + "." + iopts.format
logging.debug("Print image to `{0}` {1}".format(image_name, iopts))
plt.savefig(image_name, dpi=iopts.dpi)
plt.rcdefaults()
def main(tx=None):
"""
%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.
"""
p = OptionParser(main.__doc__)
p.add_option("--outgroup", help="Root the tree using the outgroup. " + \
"Use comma to separate multiple taxa.")
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]")
opts, args, iopts = set_image_options(p, figsize="8x6")
if len(args) != 1:
sys.exit(not p.print_help())
datafile, = args
outgroup = None
if opts.outgroup:
outgroup = opts.outgroup.split(",")
pf = datafile.rsplit(".", 1)[0]
if tx:
pf = "demo"
else:
tx = open(datafile).read()
logging.debug("Load tree file `{0}`.".format(datafile))
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
draw_tree(root, tx, rmargin=opts.rmargin,
outgroup=outgroup, gffdir=opts.gffdir, sizes=opts.sizes)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = pf + "." + iopts.format
logging.debug("Print image to `{0}` {1}".format(image_name, iopts))
plt.savefig(image_name, dpi=iopts.dpi)
plt.rcdefaults()
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.style,
proportional=proportional,
sampleN=opts.sample,
baseticks=True,
stripNames=opts.stripNames)
# 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()
logging.debug("Print image to `{0}` {1}".format(image_name, iopts))
plt.savefig(image_name, dpi=iopts.dpi)
plt.rcdefaults()
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 = set_image_options(p, args, figsize="8x5")
if len(args) != 2:
sys.exit(not p.print_help())
fastafile, chr = args
window, shift = check_window_options(opts)
stacks = opts.stacks.split(",")
heatmaps = opts.heatmaps.split(",")
stackbeds = [x + ".bed" for x in stacks]
heatmapbeds = [x + ".bed" for x in heatmaps]
stackbins = get_binfiles(stackbeds, fastafile, shift)
heatmapbins = get_binfiles(heatmapbeds, fastafile, shift)
window, shift = check_window_options(opts)
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
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 = clen / shift
if clen % shift:
nbins += 1
owindow = clen / 100
if owindow > window:
window = owindow / shift * shift
stackplot(ax, stackbins, nbins, palette, chr, window, shift)
root.text(xx + inner, yy + yinterval - 2 * inner, cc, va="top")
# 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
logging.debug("Print image to `{0}` {1}".format(image_name, iopts))
plt.savefig(image_name, dpi=iopts.dpi)
plt.rcdefaults()
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 = set_image_options(p, args, figsize="8x8")
if len(args) != 1:
sys.exit(not p.print_help())
fastafile, = args
top = opts.top
window, shift = check_window_options(opts)
switch = opts.switch
if switch:
switch = DictFile(opts.switch)
bedfiles = [x + ".bed" for x in opts.stacks.split(",")]
binfiles = get_binfiles(bedfiles, fastafile, shift)
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])
max_len = s
# 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
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
logging.debug("Print image to `{0}` {1}".format(image_name, iopts))
plt.savefig(image_name, dpi=iopts.dpi)
plt.rcdefaults()
def main():
p = OptionParser(__doc__)
p.add_option("--groups", default=False, action="store_true",
help="The first row contains group info [default: %default]")
p.add_option("--rowgroups", help="Row groupings [default: %default]")
p.add_option("--horizontalbar", default=False, action="store_true",
help="Horizontal color bar [default: vertical]")
p.add_option("--cmap", default="jet",
help="Use this color map [default: %default]")
opts, args, iopts = set_image_options(p, figsize="8x8")
if len(args) != 1:
sys.exit(not p.print_help())
datafile, = args
pf = datafile.rsplit(".", 1)[0]
rowgroups = opts.rowgroups
groups, rows, cols, data = parse_csv(datafile, vmin=1, groups=opts.groups)
cols = [x.replace("ay ", "") for x in cols]
if rowgroups:
fp = open(rowgroups)
rgroups = []
for row in fp:
a, b = row.split()
irows = [rows.index(x) for x in b.split(",")]
rgroups.append((a, min(irows), max(irows)))
plt.rcParams["axes.linewidth"] = 0
xstart = .18
fig = plt.figure(1, (iopts.w, iopts.h))
root = fig.add_axes([0, 0, 1, 1])
ax = fig.add_axes([xstart, .15, .7, .7])
default_cm = cm.get_cmap(opts.cmap)
im = ax.matshow(data, cmap=default_cm, norm=LogNorm(vmin=1, vmax=10000))
nrows, ncols = len(rows), len(cols)
xinterval = .7 / ncols
yinterval = .7 / max(nrows, ncols)
plt.xticks(range(ncols), cols, rotation=45, size=10, ha="center")
plt.yticks(range(nrows), rows, size=10)
for x in ax.get_xticklines() + ax.get_yticklines():
x.set_visible(False)
ax.set_xlim(-.5, ncols - .5)
t = [1, 10, 100, 1000, 10000]
pad = .06
if opts.horizontalbar:
ypos = .5 * (1 - nrows * yinterval) - pad
axcolor = fig.add_axes([.3, ypos, .4, .02])
orientation = "horizontal"
else:
axcolor = fig.add_axes([.9, .3, .02, .4])
orientation = "vertical"
fig.colorbar(im, cax=axcolor, ticks=t, format=_("%d"), orientation=orientation)
if groups:
groups = [(key, len(list(nn))) for key, nn in groupby(groups)]
yy = .5 + .5 * nrows / ncols * .7 + .06
e = .005
sep = -.5
for k, kl in groups:
# Separator in the array area
sep += kl
ax.plot([sep, sep], [-.5, nrows - .5], "w-", lw=2)
# Group labels on the top
kl *= xinterval
root.plot([xstart + e, xstart + kl - e], [yy, yy], "-", color="gray", lw=2)
root.text(xstart + .5 * kl, yy + e, k, ha="center", color="gray")
xstart += kl
if rowgroups:
from jcvi.graphics.glyph import TextCircle
xpos = .04
tip = .015
assert rgroups
ystart = 1 - .5 * (1 - nrows * yinterval)
for gname, start, end in rgroups:
start = ystart - start * yinterval
end = ystart - (end + 1) * yinterval
start -= tip / 3
end += tip / 3
# Bracket the groups
root.plot((xpos, xpos + tip), (start, start), "k-", lw=2)
root.plot((xpos, xpos), (start, end), "k-", lw=2)
root.plot((xpos, xpos + tip), (end, end), "k-", lw=2)
TextCircle(root, xpos, .5 * (start + end), gname)
root.set_xlim(0, 1)
root.set_ylim(0, 1)
root.set_axis_off()
image_name = pf + "." + opts.cmap + "." + iopts.format
logging.debug("Print image to `{0}` {1}".format(image_name, iopts))
plt.savefig(image_name, dpi=iopts.dpi)
plt.rcdefaults()
image_name = op.splitext(blastfile)[0] + "." + opts.format
plt.rcParams["xtick.major.pad"] = 16
plt.rcParams["ytick.major.pad"] = 16
# Fix the width
xsize, ysize = qsizes.totalsize, ssizes.totalsize
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.style, proportional=proportional, sampleN=opts.sample,
baseticks=True, stripNames=opts.stripNames)
# 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()
logging.debug("Print image to `{0}` {1}".format(image_name, iopts))
plt.savefig(image_name, dpi=iopts.dpi)
plt.rcdefaults()
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 = set_image_options(p, 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)
bedfiles = [x + ".bed" for x in opts.stacks.split(",")]
binfiles = get_binfiles(bedfiles, fastafile, shift, 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])
max_len = s
# 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
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
logging.debug("Print image to `{0}` {1}".format(image_name, iopts))
plt.savefig(image_name, dpi=iopts.dpi)
plt.rcdefaults()
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 = set_image_options(p, 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 = [x + ".bed" for x in stacks]
heatmapbeds = [x + ".bed" for x in heatmaps]
stackbins = get_binfiles(stackbeds, fastafile, shift, subtract)
heatmapbins = get_binfiles(heatmapbeds, fastafile, shift, 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
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 = clen / shift
if clen % shift:
nbins += 1
owindow = clen / 100
if owindow > window:
window = owindow / shift * shift
stackplot(ax, stackbins, nbins, palette, chr, window, shift)
root.text(xx + inner, yy + yinterval - 2 * inner, cc, va="top")
# 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
logging.debug("Print image to `{0}` {1}".format(image_name, iopts))
plt.savefig(image_name, dpi=iopts.dpi)
plt.rcdefaults()