Mercurial > repos > mheinzl > fsd
diff fsd.py @ 17:2e517a54eedc draft
planemo upload for repository https://github.com/monikaheinzl/duplexanalysis_galaxy/tree/master/tools/fsd commit b8a2f7b7615b2bcd3b602027af31f4e677da94f6-dirty
author | mheinzl |
---|---|
date | Tue, 02 Apr 2019 05:10:09 -0400 |
parents | 6bd9ef49d013 |
children | c825a29a7d9f |
line wrap: on
line diff
--- a/fsd.py Mon Oct 08 05:50:18 2018 -0400 +++ b/fsd.py Tue Apr 02 05:10:09 2019 -0400 @@ -45,19 +45,17 @@ def compare_read_families(argv): + parser = make_argparser() args = parser.parse_args(argv[1:]) - firstFile = args.inputFile1 name1 = args.inputName1 - secondFile = args.inputFile2 name2 = args.inputName2 thirdFile = args.inputFile3 name3 = args.inputName3 fourthFile = args.inputFile4 name4 = args.inputName4 - title_file = args.output_tabular title_file2 = args.output_pdf @@ -90,24 +88,32 @@ data_array_list.append(file1) legend = "\n\n\n{}".format(name1) - plt.text(0.1, 0.11, legend, size=12, transform=plt.gcf().transFigure) - legend1 = "singletons:\nabsolute nr.\n{:,}".format(numpy.bincount(data1)[1]) - plt.text(0.4, 0.11, legend1, size=12, transform=plt.gcf().transFigure) + plt.text(0.05, 0.11, legend, size=10, transform=plt.gcf().transFigure) + legend1 = "singletons:\nnr. of tags\n{:,}".format(numpy.bincount(data1)[1]) + plt.text(0.32, 0.11, legend1, size=10, transform=plt.gcf().transFigure) - legend3 = "rel. freq\n{:.3f}".format(float(numpy.bincount(data1)[1]) / len(data1)) - plt.text(0.5, 0.11, legend3, size=12, transform=plt.gcf().transFigure) + legend3 = "freq. of tags\n{:.3f}".format(float(numpy.bincount(data1)[1]) / len(data1)) + plt.text(0.41, 0.11, legend3, size=10, transform=plt.gcf().transFigure) + + legend3b = "PE reads\n{:.3f}".format(float(numpy.bincount(data1)[1]) / sum(integers)) + plt.text(0.5, 0.11, legend3b, size=10, transform=plt.gcf().transFigure) - legend4 = "family size > 20:\nabsolute nr.\n{:,}".format(numpy.bincount(data1)[len(numpy.bincount(data1)) - 1].astype(int)) - plt.text(0.6, 0.11, legend4, size=12, transform=plt.gcf().transFigure) + legend4 = "family size > 20:\nnr. of tags\n{:,} ({:.3f})".format(numpy.bincount(data1)[len(numpy.bincount(data1)) - 1].astype(int), float(numpy.bincount(data1)[len(numpy.bincount(data1)) - 1]) / len(data1)) + plt.text(0.58, 0.11, legend4, size=10, transform=plt.gcf().transFigure) + + legend5 = "PE reads\n{:,} ({:.3f})".format(sum(integers[integers > 20]), float(sum(integers[integers > 20])) / sum(integers)) + plt.text(0.70, 0.11, legend5, size=10, transform=plt.gcf().transFigure) - legend5 = "rel. freq\n{:.3f}".format(float(numpy.bincount(data1)[len(numpy.bincount(data1)) - 1]) / len(data1)) - plt.text(0.7, 0.11, legend5, size=12, transform=plt.gcf().transFigure) + legend6 = "total nr. of\ntags\n{:,}".format(len(data1)) + plt.text(0.82, 0.11, legend6, size=10, transform=plt.gcf().transFigure) - legend6 = "total length\n{:,}".format(len(data1)) - plt.text(0.8, 0.11, legend6, size=12, transform=plt.gcf().transFigure) + legend6b = "PE reads\n{:,}".format(sum(integers)) + plt.text(0.89, 0.11, legend6b, size=10, transform=plt.gcf().transFigure) if secondFile != str(None): file2 = readFileReferenceFree(secondFile) + integers2 = numpy.array(file2[:, 0]).astype(int) # keep original family sizes + data2 = numpy.asarray(file2[:, 0]).astype(int) bigFamilies2 = numpy.where(data2 > 20)[0] data2[bigFamilies2] = 22 @@ -117,25 +123,34 @@ label.append(name2) data_array_list.append(file2) - plt.text(0.1, 0.09, name2, size=12, transform=plt.gcf().transFigure) + plt.text(0.05, 0.09, name2, size=10, transform=plt.gcf().transFigure) legend1 = "{:,}".format(numpy.bincount(data2)[1]) - plt.text(0.4, 0.09, legend1, size=12, transform=plt.gcf().transFigure) + plt.text(0.32, 0.09, legend1, size=10, transform=plt.gcf().transFigure) legend3 = "{:.3f}".format(float(numpy.bincount(data2)[1]) / len(data2)) - plt.text(0.5, 0.09, legend3, size=12, transform=plt.gcf().transFigure) + plt.text(0.41, 0.09, legend3, size=10, transform=plt.gcf().transFigure) + + legend3b = "{:.3f}".format(float(numpy.bincount(data2)[1]) / sum(integers2)) + plt.text(0.5, 0.09, legend3b, size=10, transform=plt.gcf().transFigure) - legend4 = "{:,}".format(numpy.bincount(data2)[len(numpy.bincount(data2)) - 1].astype(int)) - plt.text(0.6, 0.09, legend4, size=12, transform=plt.gcf().transFigure) + legend4 = "{:,} ({:.3f})".format( + numpy.bincount(data2)[len(numpy.bincount(data2)) - 1].astype(int), + float(numpy.bincount(data2)[len(numpy.bincount(data2)) - 1]) / len(data2)) + plt.text(0.58, 0.09, legend4, size=10, transform=plt.gcf().transFigure) - legend5 = "{:.3f}".format(float(numpy.bincount(data2)[len(numpy.bincount(data2)) - 1]) / len(data2)) - plt.text(0.7, 0.09, legend5, size=12, transform=plt.gcf().transFigure) + legend5 = "{:,} ({:.3f})".format(sum(integers2[integers2 > 20]), float(sum(integers2[integers2 > 20])) / sum(integers2)) + plt.text(0.70, 0.09, legend5, size=10, transform=plt.gcf().transFigure) legend6 = "{:,}".format(len(data2)) - plt.text(0.8, 0.09, legend6, size=12, transform=plt.gcf().transFigure) + plt.text(0.82, 0.09, legend6, size=10, transform=plt.gcf().transFigure) + + legend6b = "{:,}".format(sum(integers2)) + plt.text(0.89, 0.09, legend6b, size=10, transform=plt.gcf().transFigure) if thirdFile != str(None): file3 = readFileReferenceFree(thirdFile) + integers3 = numpy.array(file3[:, 0]).astype(int) # keep original family sizes data3 = numpy.asarray(file3[:, 0]).astype(int) bigFamilies3 = numpy.where(data3 > 20)[0] @@ -146,25 +161,35 @@ label.append(name3) data_array_list.append(file3) - plt.text(0.1, 0.07, name3, size=12, transform=plt.gcf().transFigure) + plt.text(0.05, 0.07, name3, size=10, transform=plt.gcf().transFigure) legend1 = "{:,}".format(numpy.bincount(data3)[1]) - plt.text(0.4, 0.07, legend1, size=12, transform=plt.gcf().transFigure) + plt.text(0.32, 0.07, legend1, size=10, transform=plt.gcf().transFigure) legend3 = "{:.3f}".format(float(numpy.bincount(data3)[1]) / len(data3)) - plt.text(0.5, 0.07, legend3, size=12, transform=plt.gcf().transFigure) + plt.text(0.41, 0.07, legend3, size=10, transform=plt.gcf().transFigure) + + legend3b = "{:.3f}".format(float(numpy.bincount(data3)[1]) / sum(integers3)) + plt.text(0.5, 0.07, legend3b, size=10, transform=plt.gcf().transFigure) - legend4 = "{:,}".format(numpy.bincount(data3)[len(numpy.bincount(data3)) - 1].astype(int)) - plt.text(0.6, 0.07, legend4, size=12, transform=plt.gcf().transFigure) + legend4 = "{:,} ({:.3f})".format( + numpy.bincount(data3)[len(numpy.bincount(data3)) - 1].astype(int), + float(numpy.bincount(data3)[len(numpy.bincount(data3)) - 1]) / len(data3)) + plt.text(0.58, 0.07, legend4, size=10, transform=plt.gcf().transFigure) - legend5 = "{:.3f}".format(float(numpy.bincount(data3)[len(numpy.bincount(data3)) - 1]) / len(data3)) - plt.text(0.7, 0.07, legend5, size=12, transform=plt.gcf().transFigure) + legend5 = "{:,} ({:.3f})".format(sum(integers3[integers3 > 20]), + float(sum(integers3[integers3 > 20])) / sum(integers3)) + plt.text(0.70, 0.07, legend5, size=10, transform=plt.gcf().transFigure) legend6 = "{:,}".format(len(data3)) - plt.text(0.8, 0.07, legend6, size=12, transform=plt.gcf().transFigure) + plt.text(0.82, 0.07, legend6, size=10, transform=plt.gcf().transFigure) + + legend6b = "{:,}".format(sum(integers3)) + plt.text(0.89, 0.07, legend6b, size=10, transform=plt.gcf().transFigure) if fourthFile != str(None): file4 = readFileReferenceFree(fourthFile) + integers4 = numpy.array(file4[:, 0]).astype(int) # keep original family sizes data4 = numpy.asarray(file4[:, 0]).astype(int) @@ -176,28 +201,37 @@ label.append(name4) data_array_list.append(file4) - plt.text(0.1, 0.05, name4, size=12, transform=plt.gcf().transFigure) + plt.text(0.05, 0.05, name4, size=10, transform=plt.gcf().transFigure) legend1 = "{:,}".format(numpy.bincount(data4)[1]) - plt.text(0.4, 0.05, legend1, size=12, transform=plt.gcf().transFigure) + plt.text(0.32, 0.05, legend1, size=10, transform=plt.gcf().transFigure) - legend4 = "{:.3f}".format(float(numpy.bincount(data4)[1]) / len(data4)) - plt.text(0.5, 0.05, legend4, size=12, transform=plt.gcf().transFigure) + legend3 = "{:.3f}".format(float(numpy.bincount(data4)[1]) / len(data4)) + plt.text(0.41, 0.05, legend3, size=10, transform=plt.gcf().transFigure) + + legend3b = "{:.3f}".format(float(numpy.bincount(data4)[1]) / sum(integers4)) + plt.text(0.5, 0.05, legend3b, size=10, transform=plt.gcf().transFigure) - legend4 = "{:,}".format(numpy.bincount(data4)[len(numpy.bincount(data4)) - 1].astype(int)) - plt.text(0.6, 0.05, legend4, size=12, transform=plt.gcf().transFigure) + legend4 = "{:,} ({:.3f})".format( + numpy.bincount(data4)[len(numpy.bincount(data4)) - 1].astype(int), + float(numpy.bincount(data4)[len(numpy.bincount(data4)) - 1]) / len(data4)) + plt.text(0.58, 0.05, legend4, size=10, transform=plt.gcf().transFigure) - legend5 = "{:.3f}".format(float(numpy.bincount(data4)[len(numpy.bincount(data4)) - 1]) / len(data4)) - plt.text(0.7, 0.05, legend5, size=12, transform=plt.gcf().transFigure) + legend5 = "{:,} ({:.3f})".format(sum(integers4[integers4 > 20]), + float(sum(integers4[integers4 > 20])) / sum(integers4)) + plt.text(0.70, 0.05, legend5, size=10, transform=plt.gcf().transFigure) legend6 = "{:,}".format(len(data4)) - plt.text(0.8, 0.05, legend6, size=12, transform=plt.gcf().transFigure) + plt.text(0.82, 0.05, legend6, size=10, transform=plt.gcf().transFigure) + + legend6b = "{:,}".format(sum(integers4)) + plt.text(0.89, 0.05, legend6b, size=10, transform=plt.gcf().transFigure) maximumX = numpy.amax(numpy.concatenate(list_to_plot)) minimumX = numpy.amin(numpy.concatenate(list_to_plot)) counts = plt.hist(list_to_plot, bins=range(minimumX, maximumX + 1), stacked=False, edgecolor="black", - linewidth=1, label=label, align="left", alpha=0.7, rwidth=0.8) + linewidth=1, label=label, align="left", rwidth=0.8, alpha=0.7) ticks = numpy.arange(minimumX - 1, maximumX, 1) ticks1 = map(str, ticks) @@ -242,53 +276,71 @@ output_file.write("{}{}".format(int(sum(i)), sep)) # Family size distribution after DCS and SSCS - for dataset, data, name_file in zip(list_to_plot, data_array_list, label): + for dataset, data_o, name_file in zip(list_to_plot, data_array_list, label): maximumX = numpy.amax(dataset) minimumX = numpy.amin(dataset) - tags = numpy.array(data[:, 2]) - seq = numpy.array(data[:, 1]) + tags = numpy.array(data_o[:, 2]) + seq = numpy.array(data_o[:, 1]) data = numpy.array(dataset) - + data_o = numpy.array(data_o[:, 0]).astype(int) # find all unique tags and get the indices for ALL tags, but only once u, index_unique, c = numpy.unique(numpy.array(seq), return_counts=True, return_index=True) d = u[c > 1] # get family sizes, tag for duplicates duplTags_double = data[numpy.in1d(seq, d)] - duplTags = duplTags_double[0::2] # ab of DCS - duplTagsBA = duplTags_double[1::2] # ba of DCS + duplTags_double_o = data_o[numpy.in1d(seq, d)] - # duplTags_double_tag = tags[numpy.in1d(seq, d)] - # duplTags_double_seq = seq[numpy.in1d(seq, d)] + duplTags = duplTags_double[0::2] # ab of DCS + duplTags_o = duplTags_double_o[0::2] # ab of DCS + + duplTagsBA = duplTags_double[1::2] # ba of DCS + duplTagsBA_o = duplTags_double_o[1::2] # ba of DCS # get family sizes for SSCS with no partner ab = numpy.where(tags == "ab")[0] abSeq = seq[ab] + ab_o = data_o[ab] ab = data[ab] + ba = numpy.where(tags == "ba")[0] baSeq = seq[ba] + ba_o = data_o[ba] ba = data[ba] dataAB = ab[numpy.in1d(abSeq, d, invert=True)] + dataAB_o = ab_o[numpy.in1d(abSeq, d, invert=True)] + dataBA = ba[numpy.in1d(baSeq, d, invert=True)] + dataBA_o = ba_o[numpy.in1d(baSeq, d, invert=True)] list1 = [duplTags_double, dataAB, dataBA] # list for plotting # information for family size >= 3 dataAB_FS3 = dataAB[dataAB >= 3] + dataAB_FS3_o = dataAB_o[dataAB_o >= 3] dataBA_FS3 = dataBA[dataBA >= 3] + dataBA_FS3_o = dataBA_o[dataBA_o >= 3] ab_FS3 = ab[ab >= 3] ba_FS3 = ba[ba >= 3] + ab_FS3_o = ab_o[ab_o >= 3] + ba_FS3_o = ba_o[ba_o >= 3] duplTags_FS3 = duplTags[(duplTags >= 3) & (duplTagsBA >= 3)] # ab+ba with FS>=3 duplTags_FS3_BA = duplTagsBA[(duplTags >= 3) & (duplTagsBA >= 3)] # ba+ab with FS>=3 duplTags_double_FS3 = len(duplTags_FS3) + len(duplTags_FS3_BA) # both ab and ba strands with FS>=3 - fig = plt.figure() + # original FS + duplTags_FS3_o = duplTags_o[(duplTags_o >= 3) & (duplTagsBA_o >= 3)] # ab+ba with FS>=3 + duplTags_FS3_BA_o = duplTagsBA_o[(duplTags_o >= 3) & (duplTagsBA_o >= 3)] # ba+ab with FS>=3 + duplTags_double_FS3_o = sum(duplTags_FS3_o) + sum(duplTags_FS3_BA_o) # both ab and ba strands with FS>=3 + fig = plt.figure() plt.subplots_adjust(bottom=0.3) - counts = plt.hist(list1, bins=range(minimumX, maximumX + 1), stacked=True, label=["duplex", "ab", "ba"], edgecolor="black", linewidth=1, align="left", color=["#FF0000", "#5FB404", "#FFBF00"]) + counts = plt.hist(list1, bins=range(minimumX, maximumX + 1), stacked=True, label=["duplex", "ab", "ba"], + edgecolor="black", linewidth=1, align="left", color=["#FF0000", "#5FB404", "#FFBF00"], + rwidth=0.8) # tick labels of x axis ticks = numpy.arange(minimumX - 1, maximumX, 1) ticks1 = map(str, ticks) @@ -298,33 +350,56 @@ last = counts[0][2][len(counts[0][0]) - 1] # large families plt.legend(loc='upper right', fontsize=14, bbox_to_anchor=(0.9, 1), frameon=True) - # plt.title(name1, fontsize=14) + plt.title(name_file, fontsize=14) plt.xlabel("Family size", fontsize=14) plt.ylabel("Absolute Frequency", fontsize=14) plt.margins(0.01, None) plt.grid(b=True, which="major", color="#424242", linestyle=":") # extra information beneath the plot - legend = "SSCS ab= \nSSCS ba= \nDCS (total)= \nlength of dataset=" - plt.text(0.1, 0.09, legend, size=12, transform=plt.gcf().transFigure) + legend = "SSCS ab= \nSSCS ba= \nDCS (total)= \ntotal nr. of tags=" + plt.text(0.1, 0.09, legend, size=10, transform=plt.gcf().transFigure) + + legend = "nr. of tags\n\n{:,}\n{:,}\n{:,} ({:,})\n{:,}".format(len(dataAB), len(dataBA), len(duplTags), len(duplTags_double), (len(dataAB) + len(dataBA) + len(duplTags))) + plt.text(0.23, 0.09, legend, size=10, transform=plt.gcf().transFigure) - legend = "absolute numbers\n\n{:,}\n{:,}\n{:,} ({:,})\n{:,}".format(len(dataAB), len(dataBA), len(duplTags), len(duplTags_double), (len(dataAB) + len(dataBA) + len(duplTags))) - plt.text(0.35, 0.09, legend, size=12, transform=plt.gcf().transFigure) + legend5 = "PE reads\n\n{:,}\n{:,}\n{:,} ({:,})\n{:,}".format(sum(dataAB_o), sum(dataBA_o), sum(duplTags_o), sum(duplTags_double_o), (sum(dataAB_o) + sum(dataBA_o) + sum(duplTags_o))) + plt.text(0.38, 0.09, legend5, size=10, transform=plt.gcf().transFigure) - legend = "relative frequencies\nunique\n{:.3f}\n{:.3f}\n{:.3f}\n{:,}".format(float(len(dataAB)) / (len(dataAB) + len(dataBA) + len(duplTags)), float(len(dataBA)) / (len(dataAB) + len(dataBA) + len(duplTags)), float(len(duplTags)) / (len(dataAB) + len(dataBA) + len(duplTags)), (len(dataAB) + len(dataBA) + len(duplTags))) - plt.text(0.54, 0.09, legend, size=12, transform=plt.gcf().transFigure) + legend = "rel. freq. of tags\nunique\n{:.3f}\n{:.3f}\n{:.3f}\n{:,}".format(float(len(dataAB)) / (len(dataAB) + len(dataBA) + len(duplTags)), float(len(dataBA)) / (len(dataAB) + len(dataBA) + len(duplTags)), float(len(duplTags)) / (len(dataAB) + len(dataBA) + len(duplTags)), (len(dataAB) + len(dataBA) + len(duplTags))) + plt.text(0.54, 0.09, legend, size=10, transform=plt.gcf().transFigure) legend = "total\n{:.3f}\n{:.3f}\n{:.3f} ({:.3f})\n{:,}".format(float(len(dataAB)) / (len(ab) + len(ba)), float(len(dataBA)) / (len(ab) + len(ba)), float(len(duplTags)) / (len(ab) + len(ba)), float(len(duplTags_double)) / (len(ab) + len(ba)), (len(ab) + len(ba))) - plt.text(0.64, 0.09, legend, size=12, transform=plt.gcf().transFigure) + plt.text(0.64, 0.09, legend, size=10, transform=plt.gcf().transFigure) legend1 = "\nsingletons:\nfamily size > 20:" - plt.text(0.1, 0.03, legend1, size=12, transform=plt.gcf().transFigure) + plt.text(0.1, 0.03, legend1, size=10, transform=plt.gcf().transFigure) legend4 = "{:,}\n{:,}".format(singl.astype(int), last.astype(int)) - plt.text(0.35, 0.03, legend4, size=12, transform=plt.gcf().transFigure) + plt.text(0.23, 0.03, legend4, size=10, transform=plt.gcf().transFigure) legend3 = "{:.3f}\n{:.3f}".format(singl / len(data), last / len(data)) - plt.text(0.54, 0.03, legend3, size=12, transform=plt.gcf().transFigure) + plt.text(0.64, 0.03, legend3, size=10, transform=plt.gcf().transFigure) + + legend3 = "\n\n{:,}".format(sum(data_o[data_o > 20])) + plt.text(0.38, 0.03, legend3, size=10, transform=plt.gcf().transFigure) + + legend3 = "{:.3f}\n{:.3f}".format(float(singl)/sum(data_o), float(sum(data_o[data_o > 20])) / sum(data_o)) + plt.text(0.84, 0.03, legend3, size=10, transform=plt.gcf().transFigure) + + legend = "PE reads\nunique\n{:.3f}\n{:.3f}\n{:.3f}\n{:,}".format( + float(sum(dataAB_o)) / (sum(dataAB_o) + sum(dataBA_o) + sum(duplTags_o)), + float(sum(dataBA_o)) / (sum(dataAB_o) + sum(dataBA_o) + sum(duplTags_o)), + float(sum(duplTags_o)) / (sum(dataAB_o) + sum(dataBA_o) + sum(duplTags_o)), + (sum(dataAB_o) + sum(dataBA_o) + sum(duplTags_o))) + plt.text(0.74, 0.09, legend, size=10, transform=plt.gcf().transFigure) + + legend = "total\n{:.3f}\n{:.3f}\n{:.3f} ({:.3f})\n{:,}".format( + float(sum(dataAB_o)) / (sum(ab_o) + sum(ba_o)), + float(sum(dataBA_o)) / (sum(ab_o) + sum(ba_o)), + float(sum(duplTags_o)) / (sum(ab_o) + sum(ba_o)), + float(sum(duplTags_double_o)) / (sum(ab_o) + sum(ba_o)), (sum(ab_o) + sum(ba_o))) + plt.text(0.84, 0.09, legend, size=10, transform=plt.gcf().transFigure) pdf.savefig(fig) plt.close() @@ -336,23 +411,62 @@ output_file.write("absolute frequency:{}{}\n".format(sep, count[len(count) - 1])) output_file.write("relative frequency:{}{:.3f}\n\n".format(sep, float(count[len(count) - 1]) / sum(count))) - output_file.write("{}singletons:{}{}family size > 20:\n".format(sep, sep, sep)) - output_file.write("{}absolute nr.{}rel. freq{}absolute nr.{}rel. freq{}total length\n".format(sep, sep, sep, sep, sep)) - output_file.write("{}{}{}{}{:.3f}{}{}{}{:.3f}{}{}\n\n".format(name_file, sep, singl.astype(int), sep, singl / len(data), sep, last.astype(int), sep, last / len(data), sep, len(data))) + output_file.write("{}singletons:{}{}{}family size > 20:\n".format(sep, sep, sep, sep)) + output_file.write("{}nr. of tags{}rel. freq of tags{}rel.freq of PE reads{}nr. of tags{}rel. freq of tags{}nr. of PE reads{}rel. freq of PE reads{}total nr. of tags{}total nr. of PE reads\n".format(sep, sep, sep, sep, sep, sep, sep, sep, sep)) + output_file.write("{}{}{}{}{:.3f}{}{:.3f}{}{}{}{:.3f}{}{}{}{:.3f}{}{}{}{}\n\n".format( + name_file, sep, singl.astype(int), sep, singl / len(data), sep, float(singl)/sum(data_o), sep, + last.astype(int), sep, last / len(data), sep, sum(data_o[data_o > 20]), sep, float(sum(data_o[data_o > 20])) / sum(data_o), sep, len(data), sep, sum(data_o))) # information for FS >= 1 - output_file.write("The unique frequencies were calculated from the dataset where the tags occured only once (=ab without DCS, ba without DCS)\nWhereas the total frequencies were calculated from the whole dataset (=including the DCS).\n\n") - output_file.write("FS >= 1{}{}unique:{}total:\n".format(sep, sep, sep)) - output_file.write("nr./rel. freq of ab={}{}{}{:.3f}{}{:.3f}\n".format(sep, len(dataAB), sep, float(len(dataAB)) / (len(dataAB) + len(dataBA) + len( duplTags)), sep, float(len(dataAB)) / (len(ab) + len(ba)))) - output_file.write("nr./rel. freq of ba={}{}{}{:.3f}{}{:.3f}\n".format(sep, len(dataBA), sep, float(len(dataBA)) / (len(dataBA) + len(dataBA) + len(duplTags)), sep, float(len(dataBA)) / (len(ba) + len(ba)))) - output_file.write("nr./rel. freq of DCS (total)={}{} ({}){}{:.3f}{}{:.3f} ({:.3f})\n".format(sep, len(duplTags), len(duplTags_double), sep, float(len(duplTags)) / (len(dataAB) + len(dataBA) + len(duplTags)), sep, float(len(duplTags)) / ( len(ab) + len(ba)), float(len(duplTags_double)) / (len(ab) + len(ba)))) - output_file.write("length of dataset={}{}{}{}{}{}\n".format(sep, (len(dataAB) + len(dataBA) + len(duplTags)), sep, (len(dataAB) + len(dataBA) + len(duplTags)), sep, (len(ab) + len(ba)))) + output_file.write("The unique frequencies were calculated from the dataset where the tags occured only once (=ab without DCS, ba without DCS)\n" + "Whereas the total frequencies were calculated from the whole dataset (=including the DCS).\n\n") + output_file.write("FS >= 1{}nr. of tags{}nr. of PE reads{}rel. freq of tags{}{}rel. freq of PE reads:\n".format(sep, sep, sep, sep, sep)) + output_file.write("{}{}{}unique:{}total{}unique{}total:\n".format(sep, sep, sep, sep, sep, sep)) + output_file.write("SSCS ab{}{}{}{}{}{:.3f}{}{:.3f}{}{:.3f}{}{:.3f}\n".format( + sep, len(dataAB), sep, sum(dataAB_o), sep, float(len(dataAB)) / (len(dataAB) + len(dataBA) + len(duplTags)), + sep, float(sum(dataAB_o)) / (sum(dataAB_o) + sum(dataBA_o) + sum(duplTags_o)), sep, + float(len(dataAB)) / (len(ab) + len(ba)), sep, float(sum(dataAB_o)) / (sum(ab_o) + sum(ba_o)))) + output_file.write("SSCS ba{}{}{}{}{}{:.3f}{}{:.3f}{}{:.3f}{}{:.3f}\n".format( + sep, len(dataBA), sep, sum(dataBA_o), sep, float(len(dataBA)) / (len(dataBA) + len(dataBA) + len(duplTags)), + sep, float(sum(dataBA_o)) / (sum(dataBA_o) + sum(dataBA_o) + sum(duplTags_o)), sep, float(len(dataBA)) / (len(ba) + len(ba)), + sep, float(sum(dataBA_o)) / (sum(ba_o) + sum(ba_o)))) + output_file.write("DCS (total){}{} ({}){}{} ({}){}{:.3f}{}{:.3f} ({:.3f}){}{:.3f}{}{:.3f} ({:.3f})\n".format( + sep, len(duplTags), len(duplTags_double), sep, sum(duplTags_o), sum(duplTags_double_o), sep, + float(len(duplTags)) / (len(dataAB) + len(dataBA) + len(duplTags)), sep, + float(len(duplTags)) / (len(ab) + len(ba)), float(len(duplTags_double)) / (len(ab) + len(ba)), sep, + float(sum(duplTags_o)) / (sum(dataAB_o) + sum(dataBA_o) + sum(duplTags_o)), sep, + float(sum(duplTags_o)) / (sum(ab_o) + sum(ba_o)), float(sum(duplTags_double_o)) / (sum(ab_o) + sum(ba_o)))) + output_file.write("total nr. of tags{}{}{}{}{}{}{}{}{}{}{}{}\n".format( + sep, (len(dataAB) + len(dataBA) + len(duplTags)), sep, (sum(dataAB_o) + sum(dataBA_o) + sum(duplTags_o)), sep, + (len(dataAB) + len(dataBA) + len(duplTags)), sep, (len(ab) + len(ba)), sep, + (sum(dataAB_o) + sum(dataBA_o) + sum(duplTags_o)), sep, (sum(ab_o) + sum(ba_o)))) # information for FS >= 3 - output_file.write("FS >= 3{}{}unique:{}total:\n".format(sep, sep, sep)) - output_file.write("nr./rel. freq of ab={}{}{}{:.3f}{}{:.3f}\n".format(sep, len(dataAB_FS3), sep, float(len(dataAB_FS3)) / (len(dataAB_FS3) + len(dataBA_FS3) + len(duplTags_FS3)), sep, float(len(dataAB_FS3)) / (len(ab_FS3) + len(ba_FS3)))) - output_file.write("nr./rel. freq of ba={}{}{}{:.3f}{}{:.3f}\n".format(sep, len(dataBA_FS3), sep, float(len(dataBA_FS3)) / (len(dataBA_FS3) + len(dataBA_FS3) + len(duplTags_FS3)), sep, float(len(dataBA_FS3)) / (len(ba_FS3) + len(ba_FS3)))) - output_file.write("nr./rel. freq of DCS (total)={}{} ({}){}{:.3f}{}{:.3f} ({:.3f})\n".format(sep, len(duplTags_FS3), duplTags_double_FS3, sep, float(len( duplTags_FS3)) / (len(dataBA_FS3) + len(duplTags_FS3)), sep, float(len(duplTags_FS3)) / (len(ab_FS3) + len(ba_FS3)), float(duplTags_double_FS3) / (len(ab_FS3) + len(ba_FS3)))) - output_file.write("length of dataset={}{}{}{}{}{}\n".format(sep, (len(dataAB_FS3) + len(dataBA_FS3) + len(duplTags_FS3)), sep, (len(dataAB_FS3) + len(dataBA_FS3) + len(duplTags_FS3)), sep, (len(ab_FS3) + len(ba_FS3)))) + output_file.write("\nFS >= 3{}nr. of tags{}nr. of PE reads{}rel. freq of tags{}{}rel. freq of PE reads:\n".format(sep, sep, sep, sep, sep)) + output_file.write("{}{}{}unique:{}total{}unique{}total:\n".format(sep, sep, sep, sep, sep, sep)) + output_file.write("SSCS ab{}{}{}{}{}{:.3f}{}{:.3f}{}{:.3f}{}{:.3f}\n".format( + sep, len(dataAB_FS3), sep, sum(dataAB_FS3_o), sep, + float(len(dataAB_FS3)) / (len(dataAB_FS3) + len(dataBA_FS3) + len(duplTags_FS3)), sep, + float(len(dataAB_FS3)) / (len(dataBA_FS3) + len(dataBA_FS3) + duplTags_double_FS3), + sep, float(sum(dataAB_FS3_o)) / (sum(dataAB_FS3_o) + sum(dataBA_FS3_o) + sum(duplTags_FS3_o)), + sep, float(sum(dataAB_FS3_o)) / (sum(dataBA_FS3_o) + sum(dataBA_FS3_o) + duplTags_double_FS3_o))) + output_file.write("SSCS ba{}{}{}{}{}{:.3f}{}{:.3f}{}{:.3f}{}{:.3f}\n".format( + sep, len(dataBA_FS3), sep, sum(dataBA_FS3_o), sep, + float(len(dataBA_FS3)) / (len(dataBA_FS3) + len(dataBA_FS3) + len(duplTags_FS3)), + sep, float(len(dataBA_FS3)) / (len(dataBA_FS3) + len(dataBA_FS3) + duplTags_double_FS3), + sep, float(sum(dataBA_FS3_o)) / (sum(dataBA_FS3_o) + sum(dataBA_FS3_o) + sum(duplTags_FS3_o)), + sep, float(sum(dataBA_FS3_o)) / (sum(dataBA_FS3_o) + sum(dataBA_FS3_o) + duplTags_double_FS3_o))) + output_file.write("DCS (total){}{} ({}){}{} ({}){}{:.3f}{}{:.3f} ({:.3f}){}{:.3f}{}{:.3f} ({:.3f})\n".format( + sep, len(duplTags_FS3), duplTags_double_FS3, sep, sum(duplTags_FS3_o), duplTags_double_FS3_o, sep, + float(len(duplTags_FS3)) / (len(dataAB_FS3) + len(dataBA_FS3) + len(duplTags_FS3)), sep, + float(len(duplTags_FS3)) / (len(dataAB_FS3) + len(dataBA_FS3) + duplTags_double_FS3), + float(duplTags_double_FS3) / (len(dataAB_FS3) + len(dataBA_FS3) + duplTags_double_FS3), + sep, float(sum(duplTags_FS3_o)) / (sum(dataAB_FS3_o) + sum(dataBA_FS3_o) + sum(duplTags_FS3_o)), sep, + float(sum(duplTags_FS3_o)) / (sum(dataAB_FS3_o) + sum(dataBA_FS3_o) + duplTags_double_FS3_o), + float(duplTags_double_FS3_o) / (sum(dataAB_FS3_o) + sum(dataBA_FS3_o) + duplTags_double_FS3_o))) + output_file.write("total nr. of tags{}{}{}{}{}{}{}{}{}{}{}{}\n".format( + sep, (len(dataAB_FS3) + len(dataBA_FS3) + len(duplTags_FS3)), sep, (sum(dataAB_FS3_o) + sum(dataBA_FS3_o) + sum(duplTags_FS3_o)), + sep, (len(dataAB_FS3) + len(dataBA_FS3) + len(duplTags_FS3)), sep, (len(dataAB_FS3) + len(dataBA_FS3) + duplTags_double_FS3), + sep, (sum(dataAB_FS3_o) + sum(dataBA_FS3_o) + sum(duplTags_FS3_o)), sep, (sum(dataAB_FS3_o) + sum(dataBA_FS3_o) + duplTags_double_FS3_o))) output_file.write("\nValues from family size distribution\n") output_file.write("{}duplex{}ab{}ba{}sum\n".format(sep, sep, sep, sep))