def save_tick_x(results_dirpath, tick_x):
json_fpath = json_saver.save_tick_x(results_dirpath, tick_x)
if json_fpath:
append(results_dirpath, json_fpath, "tickX")
def do(ref_fpath, contigs_fpaths, output_dirpath, json_output_dir, results_dir):
logger.print_timestamp()
logger.main_info("Running Basic statistics processor...")
if not os.path.isdir(output_dirpath):
os.mkdir(output_dirpath)
reference_length = None
if ref_fpath:
reference_length = sum(fastaparser.get_lengths_from_fastafile(ref_fpath))
reference_GC, reference_GC_distribution = GC_content(ref_fpath)
logger.info(" Reference genome:")
logger.info(
" "
+ os.path.basename(ref_fpath)
+ ", Reference length = "
+ str(reference_length)
+ ", Reference GC % = "
+ "%.2f" % reference_GC
)
elif qconfig.estimated_reference_size:
reference_length = qconfig.estimated_reference_size
logger.info(" Estimated reference length = " + str(reference_length))
if reference_length:
# Saving the reference in JSON
if json_output_dir:
json_saver.save_reference_length(json_output_dir, reference_length)
# Saving for an HTML report
if qconfig.html_report:
from libs.html_saver import html_saver
html_saver.save_reference_length(results_dir, reference_length)
logger.info(" Contig files: ")
lists_of_lengths = []
numbers_of_Ns = []
for id, contigs_fpath in enumerate(contigs_fpaths):
assembly_label = qutils.label_from_fpath(contigs_fpath)
logger.info(" " + qutils.index_to_str(id) + assembly_label)
# lists_of_lengths.append(fastaparser.get_lengths_from_fastafile(contigs_fpath))
list_of_length = []
number_of_Ns = 0
for (name, seq) in fastaparser.read_fasta(contigs_fpath):
list_of_length.append(len(seq))
number_of_Ns += seq.count("N")
lists_of_lengths.append(list_of_length)
numbers_of_Ns.append(number_of_Ns)
num_contigs = max([len(list_of_length) for list_of_length in lists_of_lengths])
multiplicator = 1
if num_contigs >= (qconfig.max_points * 2):
import math
multiplicator = int(num_contigs / qconfig.max_points)
max_points = num_contigs / multiplicator
lists_of_lengths = [sorted(list, reverse=True) for list in lists_of_lengths]
corr_lists_of_lengths = [
[
sum(list_of_length[((i - 1) * multiplicator) : (i * multiplicator)])
for i in range(1, max_points)
if (i * multiplicator) < len(list_of_length)
]
for list_of_length in lists_of_lengths
]
for num_list in range(len(corr_lists_of_lengths)):
last_index = len(corr_lists_of_lengths[num_list])
corr_lists_of_lengths[num_list].append(sum(lists_of_lengths[num_list][last_index * multiplicator :]))
else:
corr_lists_of_lengths = lists_of_lengths
# saving lengths to JSON
if json_output_dir:
json_saver.save_contigs_lengths(json_output_dir, contigs_fpaths, corr_lists_of_lengths)
json_saver.save_tick_x(json_output_dir, multiplicator)
if qconfig.html_report:
from libs.html_saver import html_saver
html_saver.save_contigs_lengths(results_dir, contigs_fpaths, corr_lists_of_lengths)
html_saver.save_tick_x(results_dir, multiplicator)
########################################################################
logger.info(" Calculating N50 and L50...")
list_of_GC_distributions = []
largest_contig = 0
import N50
for id, (contigs_fpath, lengths_list, number_of_Ns) in enumerate(
itertools.izip(contigs_fpaths, lists_of_lengths, numbers_of_Ns)
):
report = reporting.get(contigs_fpath)
n50, l50 = N50.N50_and_L50(lengths_list)
ng50, lg50 = None, None
if reference_length:
ng50, lg50 = N50.NG50_and_LG50(lengths_list, reference_length)
n75, l75 = N50.N50_and_L50(lengths_list, 75)
ng75, lg75 = None, None
if reference_length:
ng75, lg75 = N50.NG50_and_LG50(lengths_list, reference_length, 75)
total_length = sum(lengths_list)
total_GC, GC_distribution = GC_content(contigs_fpath, skip=qconfig.no_gc)
list_of_GC_distributions.append(GC_distribution)
logger.info(
" "
+ qutils.index_to_str(id)
+ qutils.label_from_fpath(contigs_fpath)
+ ", N50 = "
+ str(n50)
+ ", L50 = "
+ str(l50)
+ ", Total length = "
+ str(total_length)
+ ", GC % = "
+ ("%.2f" % total_GC if total_GC is not None else "undefined")
+ ", # N's per 100 kbp = "
+ " %.2f" % (float(number_of_Ns) * 100000.0 / float(total_length))
if total_length != 0
else "undefined"
)
report.add_field(reporting.Fields.N50, n50)
report.add_field(reporting.Fields.L50, l50)
if reference_length and not qconfig.is_combined_ref:
report.add_field(reporting.Fields.NG50, ng50)
report.add_field(reporting.Fields.LG50, lg50)
report.add_field(reporting.Fields.N75, n75)
report.add_field(reporting.Fields.L75, l75)
if reference_length and not qconfig.is_combined_ref:
report.add_field(reporting.Fields.NG75, ng75)
report.add_field(reporting.Fields.LG75, lg75)
report.add_field(reporting.Fields.CONTIGS, len(lengths_list))
if lengths_list:
report.add_field(reporting.Fields.LARGCONTIG, max(lengths_list))
largest_contig = max(largest_contig, max(lengths_list))
report.add_field(reporting.Fields.TOTALLEN, total_length)
if not qconfig.is_combined_ref:
report.add_field(reporting.Fields.GC, ("%.2f" % total_GC if total_GC is not None else None))
report.add_field(reporting.Fields.UNCALLED, number_of_Ns)
report.add_field(
reporting.Fields.UNCALLED_PERCENT, ("%.2f" % (float(number_of_Ns) * 100000.0 / float(total_length)))
)
if ref_fpath:
report.add_field(reporting.Fields.REFLEN, int(reference_length))
if not qconfig.is_combined_ref:
report.add_field(reporting.Fields.REFGC, "%.2f" % reference_GC)
elif reference_length:
report.add_field(reporting.Fields.ESTREFLEN, int(reference_length))
import math
qconfig.min_difference = math.ceil((largest_contig / 1000) / 600) # divide on height of plot
if json_output_dir:
json_saver.save_GC_info(json_output_dir, contigs_fpaths, list_of_GC_distributions)
if qconfig.html_report and not qconfig.is_combined_ref:
from libs.html_saver import html_saver
html_saver.save_GC_info(results_dir, contigs_fpaths, list_of_GC_distributions)
import plotter
########################################################################
# Drawing Nx and NGx plots...
plotter.Nx_plot(
results_dir,
num_contigs > qconfig.max_points,
contigs_fpaths,
lists_of_lengths,
output_dirpath + "/Nx_plot",
"Nx",
[],
json_output_dir=json_output_dir,
)
if reference_length and not qconfig.is_combined_ref:
plotter.Nx_plot(
results_dir,
num_contigs > qconfig.max_points,
contigs_fpaths,
lists_of_lengths,
output_dirpath + "/NGx_plot",
"NGx",
[reference_length for i in range(len(contigs_fpaths))],
json_output_dir=json_output_dir,
)
if qconfig.draw_plots:
########################################################################import plotter
# Drawing cumulative plot...
plotter.cumulative_plot(
ref_fpath, contigs_fpaths, lists_of_lengths, output_dirpath + "/cumulative_plot", "Cumulative length"
)
if not qconfig.is_combined_ref:
########################################################################
# Drawing GC content plot...
list_of_GC_distributions_with_ref = list_of_GC_distributions
if ref_fpath:
list_of_GC_distributions_with_ref.append(reference_GC_distribution)
plotter.GC_content_plot(
ref_fpath, contigs_fpaths, list_of_GC_distributions_with_ref, output_dirpath + "/GC_content_plot"
)
logger.main_info("Done.")
def save_tick_x(results_dirpath, tick_x):
json_fpath = json_saver.save_tick_x(results_dirpath, tick_x)
if json_fpath:
append(results_dirpath, json_fpath, 'tickX')
def do(ref_fpath, contigs_fpaths, output_dirpath, json_output_dir, results_dir):
logger.print_timestamp()
logger.main_info("Running Basic statistics processor...")
if not os.path.isdir(output_dirpath):
os.mkdir(output_dirpath)
reference_length = None
if ref_fpath:
reference_length = sum(fastaparser.get_lengths_from_fastafile(ref_fpath))
reference_GC, reference_GC_distribution = GC_content(ref_fpath)
logger.info(' Reference genome:')
logger.info(' ' + os.path.basename(ref_fpath) + ', Reference length = ' + str(reference_length) + ', Reference GC % = ' + '%.2f' % reference_GC)
elif qconfig.estimated_reference_size:
reference_length = qconfig.estimated_reference_size
logger.info(' Estimated reference length = ' + str(reference_length))
if reference_length:
# Saving the reference in JSON
if json_output_dir:
json_saver.save_reference_length(json_output_dir, reference_length)
# Saving for an HTML report
if qconfig.html_report:
from libs.html_saver import html_saver
html_saver.save_reference_length(results_dir, reference_length)
logger.info(' Contig files: ')
lists_of_lengths = []
numbers_of_Ns = []
for id, contigs_fpath in enumerate(contigs_fpaths):
assembly_label = qutils.label_from_fpath(contigs_fpath)
logger.info(' ' + qutils.index_to_str(id) + assembly_label)
#lists_of_lengths.append(fastaparser.get_lengths_from_fastafile(contigs_fpath))
list_of_length = []
number_of_Ns = 0
for (name, seq) in fastaparser.read_fasta(contigs_fpath):
list_of_length.append(len(seq))
number_of_Ns += seq.count('N')
lists_of_lengths.append(list_of_length)
numbers_of_Ns.append(number_of_Ns)
num_contigs = max([len(list_of_length) for list_of_length in lists_of_lengths])
multiplicator = 1
if num_contigs >= (qconfig.max_points*2):
import math
multiplicator = int(num_contigs/qconfig.max_points)
max_points = num_contigs/multiplicator
lists_of_lengths = [sorted(list, reverse=True) for list in lists_of_lengths]
corr_lists_of_lengths = [[sum(list_of_length[((i-1)*multiplicator):(i*multiplicator)]) for i in range(1, max_points)
if (i*multiplicator) < len(list_of_length)] for list_of_length in lists_of_lengths]
for num_list in range(len(corr_lists_of_lengths)):
last_index = len(corr_lists_of_lengths[num_list])
corr_lists_of_lengths[num_list].append(sum(lists_of_lengths[num_list][last_index*multiplicator:]))
else:
corr_lists_of_lengths = lists_of_lengths
# saving lengths to JSON
if json_output_dir:
json_saver.save_contigs_lengths(json_output_dir, contigs_fpaths, corr_lists_of_lengths)
json_saver.save_tick_x(json_output_dir, multiplicator)
if qconfig.html_report:
from libs.html_saver import html_saver
html_saver.save_contigs_lengths(results_dir, contigs_fpaths, corr_lists_of_lengths)
html_saver.save_tick_x(results_dir, multiplicator)
########################################################################
logger.info(' Calculating N50 and L50...')
list_of_GC_distributions = []
largest_contig = 0
import N50
for id, (contigs_fpath, lengths_list, number_of_Ns) in enumerate(itertools.izip(contigs_fpaths, lists_of_lengths, numbers_of_Ns)):
report = reporting.get(contigs_fpath)
n50, l50 = N50.N50_and_L50(lengths_list)
ng50, lg50 = None, None
if reference_length:
ng50, lg50 = N50.NG50_and_LG50(lengths_list, reference_length)
n75, l75 = N50.N50_and_L50(lengths_list, 75)
ng75, lg75 = None, None
if reference_length:
ng75, lg75 = N50.NG50_and_LG50(lengths_list, reference_length, 75)
total_length = sum(lengths_list)
total_GC, GC_distribution = GC_content(contigs_fpath, skip=qconfig.no_gc)
list_of_GC_distributions.append(GC_distribution)
logger.info(' ' + qutils.index_to_str(id) +
qutils.label_from_fpath(contigs_fpath) + \
', N50 = ' + str(n50) + \
', L50 = ' + str(l50) + \
', Total length = ' + str(total_length) + \
', GC % = ' + ('%.2f' % total_GC if total_GC is not None else 'undefined') + \
', # N\'s per 100 kbp = ' + ' %.2f' % (float(number_of_Ns) * 100000.0 / float(total_length)) if total_length != 0 else 'undefined')
report.add_field(reporting.Fields.N50, n50)
report.add_field(reporting.Fields.L50, l50)
if reference_length and not qconfig.is_combined_ref:
report.add_field(reporting.Fields.NG50, ng50)
report.add_field(reporting.Fields.LG50, lg50)
report.add_field(reporting.Fields.N75, n75)
report.add_field(reporting.Fields.L75, l75)
if reference_length and not qconfig.is_combined_ref:
report.add_field(reporting.Fields.NG75, ng75)
report.add_field(reporting.Fields.LG75, lg75)
report.add_field(reporting.Fields.CONTIGS, len(lengths_list))
if lengths_list:
report.add_field(reporting.Fields.LARGCONTIG, max(lengths_list))
largest_contig = max(largest_contig, max(lengths_list))
report.add_field(reporting.Fields.TOTALLEN, total_length)
if not qconfig.is_combined_ref:
report.add_field(reporting.Fields.GC, ('%.2f' % total_GC if total_GC is not None else None))
report.add_field(reporting.Fields.UNCALLED, number_of_Ns)
report.add_field(reporting.Fields.UNCALLED_PERCENT, ('%.2f' % (float(number_of_Ns) * 100000.0 / float(total_length))))
if ref_fpath:
report.add_field(reporting.Fields.REFLEN, int(reference_length))
if not qconfig.is_combined_ref:
report.add_field(reporting.Fields.REFGC, '%.2f' % reference_GC)
elif reference_length:
report.add_field(reporting.Fields.ESTREFLEN, int(reference_length))
import math
qconfig.min_difference = math.ceil((largest_contig/1000)/600) # divide on height of plot
if json_output_dir:
json_saver.save_GC_info(json_output_dir, contigs_fpaths, list_of_GC_distributions)
if qconfig.html_report and not qconfig.is_combined_ref:
from libs.html_saver import html_saver
html_saver.save_GC_info(results_dir, contigs_fpaths, list_of_GC_distributions)
import plotter
########################################################################
# Drawing Nx and NGx plots...
plotter.Nx_plot(results_dir, num_contigs > qconfig.max_points, contigs_fpaths, lists_of_lengths, output_dirpath + '/Nx_plot', 'Nx', [], json_output_dir=json_output_dir)
if reference_length and not qconfig.is_combined_ref:
plotter.Nx_plot(results_dir, num_contigs > qconfig.max_points, contigs_fpaths, lists_of_lengths, output_dirpath + '/NGx_plot', 'NGx',
[reference_length for i in range(len(contigs_fpaths))], json_output_dir=json_output_dir)
if qconfig.draw_plots:
########################################################################import plotter
# Drawing cumulative plot...
plotter.cumulative_plot(ref_fpath, contigs_fpaths, lists_of_lengths, output_dirpath + '/cumulative_plot', 'Cumulative length')
if not qconfig.is_combined_ref:
########################################################################
# Drawing GC content plot...
list_of_GC_distributions_with_ref = list_of_GC_distributions
if ref_fpath:
list_of_GC_distributions_with_ref.append(reference_GC_distribution)
plotter.GC_content_plot(ref_fpath, contigs_fpaths, list_of_GC_distributions_with_ref, output_dirpath + '/GC_content_plot')
logger.main_info('Done.')