def linkage_decay_from_IS(IS, plot_dir=False, **kwargs):
# Load the required data
try:
db = IS.get('raw_linkage_table')
db = db.sort_values('mm').drop_duplicates(subset=['scaffold', 'position_A', 'position_B'], keep='last')\
.sort_index().drop(columns=['mm'])
stb = IS.get('scaffold2bin')
Mdb = inStrain.genomeUtilities._add_stb(db, stb, verbose=False)
assert len(Mdb) > 0
except:
logging.error("Skipping plot 5 - you don't have all required information. You need to run inStrain genome_wide first")
traceback.print_exc()
return
# Make the plot
logging.info("Plotting plot 5")
name = 'LinkageDecay_plot.pdf'
pp = PdfPages(plot_dir + name)
for genome, mdb in Mdb.groupby('genome'):
if not plot_genome(genome, IS, **kwargs):
continue
linkage_decay_plot(mdb, title=genome)
fig = plt.gcf()
fig.set_size_inches(6, 4)
fig.tight_layout()
pp.savefig(fig)#, bbox_inches='tight')
#plt.show()
plt.close(fig)
# Save the figure
pp.close()
#plt.show()
plt.close('all')
def genome_plot_from_IS(IS, plot_dir=False, **kwargs):
# Load the required data
try:
stb = IS.get('scaffold2bin')
b2s = defaultdict(list)
for s, b in stb.items():
b2s[b].append(s)
assert len(b2s.keys()) > 0
# Load the cache
covTs = kwargs.get('covT')#, IS.get('covT'))
clonTs = kwargs.get('clonT')#, IS.get('clonT'))
raw_linkage_table = kwargs.get('raw_linkage_table')#, IS.get('raw_linkage_table'))
cumulative_snv_table = kwargs.get('cumulative_snv_table')#, IS.get('cumulative_snv_table'))
scaffold2length = IS.get('scaffold2length')
rl = IS.get_read_length()
profiled_scaffolds = set(scaffold2length.keys())
except:
logging.error("Skipping plot 2 - you don't have all required information. You need to run inStrain genome_wide first")
traceback.print_exc()
return
# Make the plot
logging.info("Plotting plot 2")
name = 'genomeWide_microdiveristy_metrics.pdf'
pp = PdfPages(plot_dir + name)
for genome, scaffolds in b2s.items():
if not plot_genome(genome, IS, **kwargs):
continue
present_scaffolds = list(set(scaffolds).intersection(set(profiled_scaffolds)))
Wdb, breaks, midpoints = load_windowed_metrics(present_scaffolds,
scaffold2length,
rl,
report_midpoints=True,
covTs=covTs, clonTs=clonTs,
raw_linkage_table=raw_linkage_table,
cumulative_snv_table=cumulative_snv_table)
if len(Wdb) == 0:
logging.debug(f"{genome} could not have windowed metrics loaded")
continue
genomeWide_microdiveristy_metrics_plot(Wdb, breaks, title=genome)
fig = plt.gcf()
fig.set_size_inches(8, 5)
fig.tight_layout()
pp.savefig(fig)#, bbox_inches='tight')
#plt.show()
plt.close(fig)
# Save the figure
pp.close()
#plt.show()
plt.close('all')
def linkage_decay_type_from_IS(IS, plot_dir=False, **kwargs):
# Load the required data
try:
# Prepare
db = IS.get('raw_linkage_table')
db = db.sort_values('mm').drop_duplicates(subset=['scaffold', 'position_A', 'position_B'], keep='last')\
.sort_index().drop(columns=['mm'])
stb = IS.get('scaffold2bin')
Mdb = inStrain.genomeUtilities._add_stb(db, stb, verbose=False)
SNdb = IS.get('SNP_mutation_types')
assert SNdb is not None
if len(SNdb) == 0:
return
SNdb.loc[:,'key'] = ["{0}:{1}".format(s, p) for s, p in zip(SNdb['scaffold'], SNdb['position'])]
k2t = SNdb.set_index('key')['mutation_type'].to_dict()
Mdb.loc[:,'link_type'] = Mdb.apply(calc_link_type, axis=1, k2t=k2t)
assert len(Mdb) > 0
except:
logging.error("Skipping plot 8 - you don't have all required information. You need to run inStrain profile_genes first")
if kwargs.get('debug', False):
traceback.print_exc()
return
# Make the plot
logging.info("Plotting plot 8")
name = 'LinkageDecay_types_plot.pdf'
pp = PdfPages(plot_dir + name)
for genome, mdb in Mdb.groupby('genome'):
if not plot_genome(genome, IS, **kwargs):
continue
db = linkage_decay_type(mdb, title=genome)
fig = plt.gcf()
fig.set_size_inches(6, 4)
fig.tight_layout()
pp.savefig(fig)#, bbox_inches='tight')
#plt.show()
plt.close(fig)
# Save the figure
pp.close()
#plt.show()
plt.close('all')
def allele_freq_plot_from_IS(IS, plot_dir=False, **kwargs):
# Load the required data
try:
if not hasattr(sns, 'histplot') and callable(getattr(sns, 'histplot')):
raise Exception("Cannot make plot 4 because your seaborn is out of date- need v0.11+")
db = IS.get('cumulative_snv_table')
if len(db) == 0:
return
db = db.sort_values('mm').drop_duplicates(subset=['scaffold', 'position'], keep='last')\
.sort_index().drop(columns=['mm'])
db = db[(db['cryptic'] == False)]
if 'allele_count' in db.columns:
db = db[db['allele_count'] >= 2]
if 'morphia' in db.columns:
db = db[db['morphia'] >= 2]
stb = IS.get('scaffold2bin')
Mdb = inStrain.genomeUtilities._add_stb(db, stb, verbose=False)
assert len(Mdb) > 0
except:
logging.error("Skipping plot 4 - you don't have all required information. You need to run inStrain genome_wide first")
traceback.print_exc()
return
# Make the plot
logging.info("Plotting plot 4")
name = 'MajorAllele_frequency_plot.pdf'
pp = PdfPages(plot_dir + name)
for genome, mdb in Mdb.groupby('genome'):
if not plot_genome(genome, IS, **kwargs):
continue
db = major_allele_freq_plot(mdb, title=genome)
fig = plt.gcf()
fig.set_size_inches(6, 4)
fig.tight_layout()
pp.savefig(fig)#, bbox_inches='tight')
#plt.show()
plt.close(fig)
# Save the figure
pp.close()
#plt.show()
plt.close('all')
def mm_plot_from_IS(IS, plot_dir=False, **kwargs):
# Load the required data
try:
Mdb = kwargs.get('GWdb', False)
assert len(Mdb) > 0
if 'mm' not in Mdb:
raise Exception(
'Plot 1 cannot be created when run with --database_mode or --skip_mm_profiling'
)
# Add the number of read-pairs
readLen = int(IS.get_read_length())
Mdb['read_length'] = readLen
Mdb['mm'] = Mdb['mm'].astype(int)
Mdb.loc[:,
'ANI_level'] = [(readLen - mm) / readLen for mm in Mdb['mm']]
except:
logging.error(
"Skipping plot 1 - you don't have all required information. You need to run inStrain genome_wide first"
)
traceback.print_exc()
return
# Make the plot
logging.info("Plotting plot 1")
name = 'CoverageAndBreadth_vs_readMismatch.pdf'
pp = PdfPages(plot_dir + name)
for genome, mdb in Mdb.groupby('genome'):
if not plot_genome(genome, IS, **kwargs):
continue
mm_plot(mdb, title=genome)
fig = plt.gcf()
fig.set_size_inches(6, 4)
pp.savefig(fig)
plt.close(fig)
# Save the figure
pp.close()
plt.close('all')
def gene_histogram_from_IS(IS, plot_dir=False, **kwargs):
# Load the required data
try:
# Prepare
db = inStrain.GeneProfile.get_gene_info(IS)
stb = IS.get('scaffold2bin')
Gdb = inStrain.genomeUtilities._add_stb(db, stb, verbose=False)
if 'clonality' in Gdb.columns:
Gdb.loc[:, 'nucl_diversity'] = 1 - Gdb['clonality']
assert len(Gdb) > 0
except:
logging.error(
"Skipping plot 9 - you don't have all required information. You need to run inStrain profile_genes first"
)
if kwargs.get('debug', False):
traceback.print_exc()
return
# Make the plot
logging.info("Plotting plot 9")
name = 'GeneHistogram_plot.pdf'
pp = PdfPages(plot_dir + name)
for genome, mdb in Gdb.groupby('genome'):
if not plot_genome(genome, IS, **kwargs):
continue
db = gene_histogram_plot(mdb, title=genome)
fig = plt.gcf()
fig.set_size_inches(8, 5)
fig.tight_layout()
pp.savefig(fig) #, bbox_inches='tight')
#plt.show()
plt.close(fig)
# Save the figure
pp.close()
#plt.show()
plt.close('all')
def ANI_dist_plot_from_IS(IS, plot_dir=False, **kwargs):
# Load the required data
try:
Mdb = prepare_read_ani_dist_plot(IS)
if len(Mdb['ANI_level'].unique()) == 1:
raise Exception(
'Plot 3 cannot be created when run with --database_mode or --skip_mm_profiling'
)
assert len(Mdb) > 0
except:
logging.error(
"Skipping plot 3 - you don't have all required information. You need to run inStrain genome_wide first"
)
traceback.print_exc()
return
# Make the plot
logging.info("Plotting plot 3")
name = 'readANI_distribution.pdf'
pp = PdfPages(plot_dir + name)
for genome, mdb in Mdb.groupby('genome'):
if not plot_genome(genome, IS, **kwargs):
continue
db = read_ani_dist_plot(mdb, title=genome)
fig = plt.gcf()
fig.set_size_inches(6, 4)
fig.tight_layout()
pp.savefig(fig) #, bbox_inches='tight')
#plt.show()
plt.close(fig)
# Save the figure
pp.close()
#plt.show()
plt.close('all')