def make_zscores(mutation, clinical, breaks, outdir):
clinical_data = util.get_clinical_data(clinical)
mut = mutation_base.prep_mutation_data(mutation, clinical_data)
cancer_type = util.get_cancer_type(mutation)
print cancer_type
structural_breaks = pd.read_csv(breaks, index_col=0)
structural_breaks = structural_breaks.astype(int)
mut_and_breaks = mut.join(structural_breaks, how='inner')
num_patients = len(mut_and_breaks)
results = []
for gene in mut_and_breaks:
if gene in ('time', 'censor', 'breaks'): # skip metadata
continue
num_mutations = mut_and_breaks[gene].sum()
if num_mutations >= MUTATION_PERCENT * num_patients:
cox_dict = analysis.do_multivariate_cox(mut_and_breaks.time,
mut_and_breaks.censor,
mut_and_breaks[gene],
mut_and_breaks[['breaks']])
cox_dict['gene'] = gene
results.append(cox_dict)
results_df = pd.DataFrame(results)
results_df = results_df.set_index('gene')
results_df.to_csv(os.path.join(outdir, cancer_type + '_mut_cox.csv'))
def make_cn_zscores(copy_number, breaks, interesting_genes=None, outdir='.'):
clinical_data = pd.read_csv(breaks, index_col=0)
cnv = pd.read_csv(copy_number, index_col=0)
cnv_by_patient = cnv.transpose()
clinical_and_cnv = cnv_by_patient.join(clinical_data, how='inner')
cancer_type = util.get_cancer_type(copy_number)
print cancer_type
if interesting_genes is not None:
relevant_genes = ('\'' + interesting_genes[
interesting_genes['Cancer Type'] == cancer_type]['Gene']).values
relevant_genes = list(relevant_genes) + ['breaks', 'time', 'censor']
print relevant_genes
clinical_and_cnv = clinical_and_cnv[relevant_genes]
clinical_and_cnv.to_csv(
os.path.join(outdir, cancer_type + '_interesting_genes_data.csv'))
return
results = []
for gene in clinical_and_cnv:
if gene in ('time', 'censor'): # skip metadata
continue
if clinical_and_cnv[gene].count() > 10:
cox_dict = analysis.do_multivariate_cox(
clinical_and_cnv.time, clinical_and_cnv.censor,
clinical_and_cnv[gene], clinical_and_cnv[['breaks']])
cox_dict['gene'] = gene
results.append(cox_dict)
results_df = pd.DataFrame(results)
results_df = results_df.set_index('gene')
results_df.to_csv(os.path.join(outdir, cancer_type + '_cn_cox.csv'))
def do_cox_models(clinical, cn_file, outdir):
cn = pd.read_csv(cn_file, sep='\t', index_col=0)
cn_by_patient = cn.transpose()
cn_by_patient = cn_by_patient.drop(['Entrez_Gene_Id'])
cn = cn_by_patient[['MYC']]
data = cn.join(clinical, how='inner')
analyses = {
'CNA only': [age_r, 'her2_0', 'combined_er_pr', 'stage_0', 'stage_1'],
}
results = pd.DataFrame()
pp = pprint.PrettyPrinter(indent=2)
for g in cn:
for name, a in analyses.iteritems():
cox_dict = analysis.do_multivariate_cox(data.Time,
data.Censor,
data[g],
data[a],
float_vars=True)
cox_dict['gene'] = name + ' ' + g
results = results.append(cox_dict, ignore_index=True)
results = results.set_index('gene')
results.T.to_csv(os.path.join(outdir, 'breast_analysis.csv'))
def do_cox_models(clinical, cn_file, mut_file, outdir):
cn = pd.read_csv(cn_file)
cn_by_patient = cn.transpose()
cn_by_patient = cn_by_patient.drop(['Chromosome', 'Location'])
cn_by_patient.columns = cn_by_patient.loc['Symbol']
cn = cn_by_patient[['\'MYC']]
mut = mutation_base.prep_mutation_data(mut_file, clinical)
p53_mut = mut[['\'TP53']]
p53_mut.columns = ['TP53']
data = cn.join(clinical, how='inner')
data = data.join(p53_mut, how='inner')
analyses = {
'CNA only': [age_r, 'her2_0', 'combined_er_pr', 'stage_0', 'stage_1'],
'CNA + P53':
['TP53', age_r, 'her2_0', 'combined_er_pr', 'stage_0', 'stage_1']
}
results = pd.DataFrame()
pp = pprint.PrettyPrinter(indent=2)
for g in cn:
for name, a in analyses.iteritems():
cox_dict = analysis.do_multivariate_cox(data.time,
data.censor,
data[g],
data[a],
float_vars=True)
cox_dict['gene'] = name + ' ' + g
results = results.append(cox_dict, ignore_index=True)
cox_dict = analysis.do_multivariate_cox(data.time,
data.censor,
data['TP53'],
data[analyses['CNA only']],
float_vars=True)
cox_dict['gene'] = 'TP53 mut'
results = results.append(cox_dict, ignore_index=True)
results = results.set_index('gene')
results.T.to_csv(os.path.join(outdir, 'breast_analysis.csv'))
def calculate_cox(data, gene):
try:
cox_dict = analysis.do_multivariate_cox(
data.time,
data.censor,
data[gene], #'\'' + gene.split('_')[0]],
data[['TP53_mutation']],
float_vars=True)
return cox_dict
except rpy2.rinterface.RRuntimeError as e:
print 'WARN: skipped', gene, 'due to R error'
return {}
def calculate_cox(mutation, clinical_data, tumor_stage_file, outdir):
df = prep_data(mutation, clinical_data)
df = df.join(clinical_data, how='inner')
tumor_stage_data = pd.read_csv(tumor_stage_file, index_col=0)
tumor_stage_cols = [i for i in tumor_stage_data if 'group' in i]
df = df.join(tumor_stage_data[tumor_stage_cols], how='inner')
num_patients = len(df.index)
#prep output file
cancer_type = os.path.basename(mutation).split('_')[0].split('.')[0]
outfile = os.path.join(
outdir, cancer_type + '_mutation-fraction-' + str(MUTATION_PERCENT) +
'.zscores.out.csv')
header, formatstring = tumor_stage.tumor_stage_output_header_and_format(
5, tumor_stage_cols)
with open(outfile, 'w') as out:
out.write('gene,zscore,pvalue,num mutations,num patients')
out.write(header)
out.write('\n')
for gene in df:
if gene in ['time', 'censor'] + tumor_stage_cols:
continue
num_mutations = df[gene].sum()
if num_mutations >= MUTATION_PERCENT * num_patients:
analysis_data = pd.DataFrame()
analysis_data['time'] = df['time']
analysis_data['censor'] = df['censor']
analysis_data['mutated'] = df[gene].fillna(0)
analysis_data[tumor_stage_cols] = df[tumor_stage_cols]
#Do analysis!
cox_dict = analysis.do_multivariate_cox(
analysis_data['time'], analysis_data['censor'],
analysis_data['mutated'], analysis_data[tumor_stage_cols])
tumor_stage_zscores = tumor_stage.zscores_for_tumor_stage_cols(
cox_dict, tumor_stage_cols)
out.write(
formatstring.format(gene, cox_dict['var-z'],
cox_dict['var-p'], num_mutations,
cox_dict['var-n'],
*tumor_stage_zscores))
analysis_data.to_csv(
os.path.join(outdir, gene[1:] + '_data.csv'),
columns=['time', 'censor', 'mutated'] + tumor_stage_cols)
def make_zscores(copy_number, clinical_data, tumor_stage_data_dir, outdir):
cancer_type = util.get_cancer_type(copy_number)
df = pd.read_csv(copy_number)
df_by_patient = df.transpose()
df_by_patient.columns = df_by_patient.loc['Symbol']
clinical_and_cnv = df_by_patient.join(clinical_data, how='inner')
tumor_stage_data, tumor_stage_cols = tumor_stage.prep_tumor_stage_data(
tumor_stage_data_dir, cancer_type)
if tumor_stage_data is None:
return
clinical_and_cnv_and_extra = clinical_and_cnv.join(
tumor_stage_data[tumor_stage_cols], how='inner')
outfile = os.path.join(outdir, cancer_type + '_extra_clinical_zscores.csv')
header, formatstring = tumor_stage.tumor_stage_output_header_and_format(
4, tumor_stage_cols)
with open(outfile, 'w') as out:
out.write('gene,zscore,pvalue,num patients')
out.write(header)
out.write('\n')
for gene in clinical_and_cnv_and_extra:
if gene in ['time', 'censor'] + tumor_stage_cols: # skip metadata
continue
if clinical_and_cnv_and_extra[gene].count() > 10:
cox_dict = analysis.do_multivariate_cox(
clinical_and_cnv_and_extra.time,
clinical_and_cnv_and_extra.censor,
clinical_and_cnv_and_extra[gene],
clinical_and_cnv_and_extra[tumor_stage_cols])
group_zscores = tumor_stage.zscores_for_tumor_stage_cols(
cox_dict, tumor_stage_cols)
out.write(
formatstring.format(gene, cox_dict['var-z'],
cox_dict['var-p'], cox_dict['var-n'],
*group_zscores))