def plot_s(genedata: pd.DataFrame, chrom_offsets: pd.DataFrame, outpath: str,
ymin: Optional[float], ymax: Optional[float]) -> None:
if (not ymin) or (not ymax):
ymin = max(min(genedata["s"]), -1)
ymax = min(max(genedata["s"]), 1)
mh.plot_manhat(genedata,
outpath,
chrom_offsets,
"s",
title="Selection coefficient",
yname="s",
dims=(20, 6),
scale=1.5,
log=True,
named_xticks=True,
chrom_col="Scaffold",
geom="line",
ylim=(ymin, ymax))
def plot_p(genedata: pd.DataFrame, chrom_offsets: pd.DataFrame, outpath: str,
ymin: Optional[float], ymax: Optional[float]) -> None:
if (not ymin) or (not ymax):
ymin = max(min(genedata["p"]), 0)
ymax = max(genedata["p"])
mh.plot_manhat(
genedata,
outpath,
chrom_offsets,
"p",
title="Probability of selection coefficient occurring randomly",
yname="-log10(p)",
dims=(20, 6),
scale=1.5,
log=True,
named_xticks=True,
chrom_col="Scaffold",
geom="line",
ylim=(ymin, ymax))
def plot_p(genedata: pd.DataFrame, chrom_offsets: pd.DataFrame, outpath: str,
ymin: Optional[float], ymax: Optional[float]) -> None:
if (not ymin) or (not ymax):
ymin = max(min(genedata["p"]), 0)
ymax = max(genedata["p"])
mh.plot_manhat(
genedata,
outpath,
chrom_offsets,
"p",
title="Tukey test probability of\ndifference in selection coefficient",
yname="-log10(p)",
dims=(20, 6),
scale=1.5,
log=True,
named_xticks=True,
chrom_col="Scaffold",
geom="point",
ylim=(ymin, ymax),
color_col="reject")
def main():
with open("reflens.bed", "r") as inconn:
chrlens = mh.get_chrom_lens_from_bed(inconn)
print(chrlens)
gdata = mh.get_data_from_bed("answer_dens.bed", data_col_name="Density")
print(gdata)
rdata = mh.get_data_from_bed("repanswer_dens.bed", data_col_name="Density")
print(rdata)
genedata, chrom_offsets = mh.manhatify(gdata,
chrlens,
chrom_col="Scaffold",
bp_col="Start",
val_col="Value",
feature="Genes")
repdata, chrom_offsets = mh.manhatify(rdata,
chrlens,
chrom_col="Scaffold",
bp_col="Start",
val_col="Value",
feature="Repeats")
combo = mh.combine_data((genedata, repdata))
print(combo.head())
print(chrom_offsets)
mh.plot_manhat(combo,
"select_families_dens2.pdf",
chrom_offsets,
"Density",
title="Feature densities",
yname="Features per 1Mb window",
dims=(20, 6),
scale=1.5,
facet_col="Feature")
mh.plot_manhat(combo,
"select_families_dens3.pdf",
chrom_offsets,
"Density",
title="Feature densities",
yname="Features per 1Mb window",
dims=(20, 6),
scale=1.5,
color_col="Feature")
def main():
# parse all arguments
parser = argparse.ArgumentParser(
"Visualize Hi-C pairing rates as a 2-d line plot.")
parser.add_argument(
"input",
nargs='*',
help="Input file(s) generated by pairviz (default = stdin).")
parser.add_argument("-L",
"--chrlens",
help="chromosome lengths .bed file (required).",
required=True)
parser.add_argument("-o",
"--output",
help="output path (default = out.pdf).")
parser.add_argument("-t",
"--title",
help="Title of plot (default = \"Pairing Rate\").")
parser.add_argument(
"-p",
"--proportion",
help=
"If included, plot as a proportion of total reads in the region, rather than absolute (default = False).",
action="store_true")
parser.add_argument(
"-f",
"--no_fpkm",
help=
"If included, plot read counts rather than FPKM (default = False, can be combined with --proportion).",
action="store_true")
parser.add_argument("-s",
"--self",
help="Also plot self-interactions (default = False).",
action="store_true")
parser.add_argument(
"-c",
"--chromspace",
help=
"bp of space to put between chromosomes in plot (default = 5000000).")
parser.add_argument("-l",
"--log",
help="Log-scale the y-axis (default = False).",
action="store_true")
parser.add_argument("-i",
"--stdin",
help="take input from stdin along with other inputs.",
action="store_true")
parser.add_argument(
"-n",
"--name_col",
help="Name of the column to use for differentiating different runs")
parser.add_argument("-x", "--x_axis_name", help="X axis name.")
parser.add_argument("-y", "--y_axis_name", help="Y axis name.")
parser.add_argument("-N",
"--named_xticks",
help="Use chromosome names for X axis ticks.",
action="store_true")
args = parser.parse_args()
# Set all variables to defaule values, handle argument logic
chrlens_path = args.chrlens
output = "out.pdf"
inconns = []
if args.stdin or not input:
inconns.append(sys.stdin)
title = "Pairing Rate"
proportion = False
self = False
chromspace = 5000000
log = False
name_col = None
use_fpkm = True
xname = "Genome position (bp)"
yname = "Hi-C contacts"
named_xticks = False
if args.output:
output = args.output
if args.input:
for i in args.input:
inconns.append(open(i, "r"))
if args.title:
title = args.title
if args.proportion:
proportion = args.proportion
if args.self:
self = args.self
if args.chromspace:
chromspace = args.chromspace
if args.log:
log = args.log
if args.no_fpkm:
use_fpkm = False
if args.x_axis_name:
xname = args.x_axis_name
if args.y_axis_name:
yname = args.y_axis_name
if args.name_col:
name_col = args.name_col
if args.named_xticks:
named_xticks = True
# set proportion vs total hits
if not use_fpkm:
if proportion:
my_y = 'pair_prop'
alt_y = 'alt_prop'
else:
my_y = 'hits'
alt_y = 'alt_hits'
else:
if proportion:
my_y = 'pair_prop_fpkm'
alt_y = 'alt_prop_fpkm'
else:
my_y = 'pair_fpkm'
alt_y = 'alt_fpkm'
# make the combined data frame
alldatas = parse_all_data(inconns)
with open(chrlens_path, "r") as inconn:
chrlens = mh.get_chrom_lens_from_bed(inconn)
big_alldata = pd.concat(alldatas, ignore_index=True)
if name_col:
m_alldata = pd.melt(big_alldata,
id_vars=['chrom', 'start', 'end', name_col])
else:
m_alldata = pd.melt(big_alldata, id_vars=['chrom', 'start', 'end'])
if self:
mm_alldata = m_alldata[m_alldata.apply(lambda x: x['variable'] in
(my_y, alt_y),
axis=1)]
else:
mm_alldata = m_alldata[m_alldata.apply(lambda x: x['variable'] == my_y,
axis=1)]
mm_alldata['value'] = mm_alldata['value'].astype(float)
manhat_data, chroffsets = mh.manhatify(mm_alldata,
chrlens,
chrom_col="chrom",
bp_col="start",
val_col="value",
feature=name_col)
mh.plot_manhat(manhat_data,
output,
chroffsets,
"value",
title=title,
xname=xname,
yname=yname,
color_col=name_col,
log=log,
scale=1,
named_xticks=named_xticks,
chrom_col="chrom")