def main():
'''
Organization function
-setups logging
-gets inputdata
-calls plotting function
'''
settings, args = utils.get_args()
try:
utils.make_output_dir(args.outdir)
utils.init_logs(args)
args.format = nanoplotter.check_valid_format(args.format)
sources = {
"fastq": args.fastq,
"bam": args.bam,
"cram": args.cram,
"fastq_rich": args.fastq_rich,
"fastq_minimal": args.fastq_minimal,
"summary": args.summary,
"fasta": args.fasta,
"ubam": args.ubam,
}
if args.pickle:
datadf = pickle.load(open(args.pickle, 'rb'))
else:
datadf = get_input(source=[n for n, s in sources.items() if s][0],
files=[f for f in sources.values() if f][0],
threads=args.threads,
readtype=args.readtype,
combine="simple",
barcoded=args.barcoded,
huge=args.huge)
if args.store:
pickle.dump(obj=datadf,
file=open(settings["path"] + "NanoPlot-data.pickle",
'wb'))
if args.raw:
datadf.to_csv(settings["path"] + "NanoPlot-data.tsv.gz",
sep="\t",
index=False,
compression="gzip")
settings["statsfile"] = [make_stats(datadf, settings, suffix="")]
datadf, settings = filter_and_transform_data(datadf, settings)
if settings[
"filtered"]: # Bool set when filter was applied in filter_and_transform_data()
settings["statsfile"].append(
make_stats(datadf[datadf["length_filter"]],
settings,
suffix="_post_filtering"))
if args.barcoded:
barcodes = list(datadf["barcode"].unique())
plots = []
for barc in barcodes:
logging.info("Processing {}".format(barc))
settings["path"] = path.join(args.outdir,
args.prefix + barc + "_")
dfbarc = datadf[datadf["barcode"] == barc]
if len(dfbarc) > 5:
settings["title"] = barc
plots.extend(make_plots(dfbarc, settings))
else:
sys.stderr.write(
"Found barcode {} less than 5x, ignoring...\n".format(
barc))
logging.info(
"Found barcode {} less than 5 times, ignoring".format(
barc))
else:
plots = make_plots(datadf, settings)
make_report(plots, settings)
logging.info("Finished!")
except Exception as e:
logging.error(e, exc_info=True)
print("\n\n\nIf you read this then NanoPlot {} has crashed :-(".format(
__version__))
print("Please try updating NanoPlot and see if that helps...\n")
print(
"If not, please report this issue at https://github.com/wdecoster/NanoPlot/issues"
)
print(
"If you could include the log file that would be really helpful.")
print("Thanks!\n\n\n")
raise
def scatter(x,
y,
legacy,
names,
path,
plots,
color,
colormap,
settings,
stat=None,
log=False,
minvalx=0,
minvaly=0,
title=None,
xmax=None,
ymax=None):
"""->
create marginalised scatterplots and KDE plot with marginalized histograms
-> update from scatter_legacy function to utilise plotly package
- scatterplot with histogram on both axes
- kernel density plot with histograms on both axes
- hexbin not implemented yet
- pauvre plot temporarily not available
"""
logging.info(
f"NanoPlot: Creating {names[0]} vs {names[1]} plots using {x.size} reads."
)
if not contains_variance([x, y], names):
return []
plots_made = []
idx = np.random.choice(x.index, min(10000, len(x)), replace=False)
maxvalx = xmax or np.amax(x[idx])
maxvaly = ymax or np.amax(y[idx])
if plots["dot"]:
if log:
dot_plot = Plot(path=path + "_loglength_dot.html",
title=f"{names[0]} vs {names[1]} plot using dots "
"after log transformation of read lengths")
else:
dot_plot = Plot(path=path + "_dot.html",
title=f"{names[0]} vs {names[1]} plot using dots")
fig = px.scatter(x=x[idx],
y=y[idx],
marginal_x="histogram",
marginal_y="histogram",
range_x=[minvalx, maxvalx],
range_y=[minvaly, maxvaly])
fig.update_traces(marker=dict(color=color))
fig.update_yaxes(rangemode="tozero")
fig.update_xaxes(rangemode="tozero")
fig.update_layout(xaxis_title=names[0],
yaxis_title=names[1],
title=title or dot_plot.title,
title_x=0.5)
if log:
ticks = [
10**i for i in range(10) if not 10**i > 10 * (10**maxvalx)
]
fig.update_layout(xaxis=dict(tickmode='array',
tickvals=np.log10(ticks),
ticktext=ticks,
tickangle=45))
dot_plot.fig = fig
dot_plot.html = dot_plot.fig.to_html(full_html=False,
include_plotlyjs='cdn')
dot_plot.save(settings)
plots_made.append(dot_plot)
if plots["kde"]:
kde_plot = Plot(path=path + "_loglength_kde.html" if log else path +
"_kde.html",
title=f"{names[0]} vs {names[1]} kde plot")
col = hex_to_rgb_scale_0_1(color)
fig = ff.create_2d_density(x[idx],
y[idx],
point_size=3,
hist_color=col,
point_color=col,
colorscale=colormap)
fig.update_layout(xaxis_title=names[0],
yaxis_title=names[1],
title=title or kde_plot.title,
title_x=0.5,
xaxis=dict(tickangle=45))
if log:
ticks = [
10**i for i in range(10) if not 10**i > 10 * (10**maxvalx)
]
fig.update_layout(xaxis=dict(tickmode='array',
tickvals=np.log10(ticks),
ticktext=ticks,
tickangle=45))
kde_plot.fig = fig
kde_plot.html = kde_plot.fig.to_html(full_html=False,
include_plotlyjs='cdn')
kde_plot.save(settings)
plots_made.append(kde_plot)
if 1 in legacy.values():
settings, args = utils.get_args()
plots_made += scatter_legacy(x=x[idx],
y=y[idx],
names=names,
path=path,
plots=legacy,
color=color,
settings=settings,
stat=stat,
log=log,
minvalx=minvalx,
minvaly=minvaly,
title=title)
return plots_made