def _plot_stats(st, pdf):
"Generate plots and save to report PDF"
R = report.Report(pdf)
_plot_pd_bars(st.loc[st.Category == "Classification", ].copy(),
"Classification of output reads",
R,
ann=True)
_plot_pd_bars(st.loc[st.Category == "Strand", ].copy(),
"Strand of oriented reads",
R,
ann=True)
_plot_pd_bars(st.loc[st.Category == "RescueStrand", ].copy(),
"Strand of rescued reads",
R,
ann=True)
_plot_pd_bars(st.loc[st.Category == "UnclassHitNr", ].copy(),
"Number of hits in unclassified reads", R)
_plot_pd_bars(st.loc[st.Category == "RescueHitNr", ].copy(),
"Number of hits in rescued reads", R)
_plot_pd_bars(st.loc[st.Category == "RescueSegmentNr", ].copy(),
"Number of usable segments per rescued read", R)
if args.Y > 0:
_plot_pd_line(
st.loc[st.Category == "AutotuneSample", ].copy(),
"Classified reads as function of cutoff(q). Best q={:.4f}".format(
args.q),
R,
vline=args.q)
udf = st.loc[st.Category == "Unusable", ].copy()
udf.Name = np.log10(1.0 + np.array(udf.Name, dtype=float))
_plot_pd_line(udf, "Log10 length distribution of trimmed away sequences.",
R)
R.close()
def _plot_stats(st, pdf):
"Generate plots and save to report PDF"
R = report.Report(pdf)
rs = st.loc[st.Category == "Classification", ]
_plot_pd_bars(rs.copy(), "Classification of output reads", R, ann=True)
found, rescue, unusable = float(rs.loc[rs.Name == "Primers_found", ].Value), float(rs.loc[rs.Name == "Rescue", ].Value), float(rs.loc[rs.Name == "Unusable", ].Value)
total = found + rescue + unusable
found = found / total * 100
rescue = rescue / total * 100
unusable = unusable / total * 100
sys.stderr.write("-----------------------------------\n")
sys.stderr.write("Reads with two primers:\t{:.2f}%\nRescued reads:\t\t{:.2f}%\nUnusable reads:\t\t{:.2f}%\n".format(found, rescue, unusable))
sys.stderr.write("-----------------------------------\n")
_plot_pd_bars(st.loc[st.Category == "Strand", ].copy(), "Strand of oriented reads", R, ann=True)
_plot_pd_bars(st.loc[st.Category == "RescueStrand", ].copy(), "Strand of rescued reads", R, ann=True)
_plot_pd_bars(st.loc[st.Category == "UnclassHitNr", ].copy(), "Number of hits in unclassified reads", R)
_plot_pd_bars(st.loc[st.Category == "RescueHitNr", ].copy(), "Number of hits in rescued reads", R)
_plot_pd_bars(st.loc[st.Category == "RescueSegmentNr", ].copy(), "Number of usable segments per rescued read", R)
if q_bak is None:
_plot_pd_line(st.loc[st.Category == "AutotuneSample", ].copy(), "Usable bases as function of cutoff(q). Best q={:.4g}".format(args.q), R, vline=args.q)
udf = st.loc[st.Category == "Unusable", ].copy()
udf.Name = np.log10(1.0 + np.array(udf.Name, dtype=float))
_plot_pd_line(udf, "Log10 length distribution of trimmed away sequences.", R)
R.close()
scores_handle.write(str(match_dir))
scores_handle.write("\n")
output_handle.flush()
output_handle.close()
if args.u is not None:
unclass_handle.flush()
unclass_handle.close()
if args.A is not None:
scores_handle.flush()
scores_handle.close()
if args.r is not None:
plotter = report.Report(args.r)
plotter.plot_bars_simple(
{
'Classified': fwd_matches + rev_matches,
'Unclassified': unclassified
},
title="Basic statistics",
ylab="Count")
plotter.plot_histograms({'nr_hits': unclass_nr_hits},
title="Number of hits in unclassified reads",
xlab="Number of hits",
ylab="Count")
plotter.plot_bars_simple({
'+': fwd_matches,
'-': rev_matches
},