tanu/LSHTM_analysis
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

tidying script

Browse source
This commit is contained in:
Tanushree Tunstall 2020-06-25 13:12:09 +01:00
parent cdb1ea1476
commit 7032baa08d
1 changed files with 133 additions and 197 deletions
Download patch file Download diff file Expand all files Collapse all files

330
scripts/combine_afs_ors.py
View file

@ -30,41 +30,46 @@ os.chdir(homedir + '/git/LSHTM_analysis/scripts')
os.getcwd() os.getcwd()
# local import # local import
from reference_dict import my_aa_dict # CHECK DIR STRUC THERE! #from reference_dict import my_aa_dict # CHECK DIR STRUC THERE!
from reference_dict import low_3letter_dict
#======================================================================= #=======================================================================
#%% command line args #%% command line args
arg_parser = argparse.ArgumentParser() #arg_parser = argparse.ArgumentParser()
arg_parser.add_argument('-d', '--drug', help = 'drug name', default = 'pyrazinamide') #arg_parser.add_argument('-d', '--drug', help = 'drug name', default = 'pyrazinamide')
arg_parser.add_argument('-g', '--gene', help = 'gene name', default = 'pncA') # case sensitive #arg_parser.add_argument('-g', '--gene', help = 'gene name', default = 'pncA') # case sensitive
args = arg_parser.parse_args() #args = arg_parser.parse_args()
#======================================================================= #=======================================================================
#%% variable assignment: input and output #%% variable assignment: input and output
#drug = 'pyrazinamide' drug = 'pyrazinamide'
#gene = 'pncA' gene = 'pncA'
#gene_match = gene + '_p.' gene_match = gene + '_p.'
# cmd variables # cmd variables
drug = args.drug #drug = args.drug
gene = args.gene #gene = args.gene
gene_match = gene + '_p.' #gene_match = gene + '_p.'
#========== #==========
# dir # dir
#========== #==========
datadir = homedir + '/' + 'git/Data' datadir = homedir + '/' + 'git/Data'
indir = datadir + '/' + drug + '/' + 'input'
outdir = datadir + '/' + drug + '/' + 'output' outdir = datadir + '/' + drug + '/' + 'output'
#======= #=======
# input # input
#======= #=======
in_filename_snpinfo = 'ns' + gene.lower() + '_snp_info.csv'
in_filename_afor = gene.lower() + '_af_or.csv' in_filename_afor = gene.lower() + '_af_or.csv'
in_filename_afor_kin = gene.lower() + '_af_or_kinship.csv' in_filename_afor_kin = gene.lower() + '_af_or_kinship.csv'
infile0 = indir + '/' + in_filename_snpinfo
infile1 = outdir + '/' + in_filename_afor infile1 = outdir + '/' + in_filename_afor
infile2 = outdir + '/' + in_filename_afor_kin infile2 = outdir + '/' + in_filename_afor_kin
print('Input file1:', infile1 print('Input file0:', infile0
, '\nInput file1:', infile1
, '\nInput file2:', infile2 , '\nInput file2:', infile2
, '\n===================================================================') , '\n===================================================================')
@ -77,7 +82,7 @@ print('Output file:', outfile
, '\n===================================================================') , '\n===================================================================')
del(in_filename_afor, in_filename_afor_kin, datadir, outdir) del(in_filename_afor, in_filename_afor_kin, datadir, indir, outdir)
#%% end of variable assignment for input and output files #%% end of variable assignment for input and output files
#======================================================================= #=======================================================================
#%% format mutations #%% format mutations
@ -86,211 +91,142 @@ del(in_filename_afor, in_filename_afor_kin, datadir, outdir)
#======================== #========================
# read input csv files to combine # read input csv files to combine
#======================== #========================
snpinfo_df = pd.read_csv(infile0, sep = ',')
snpinfo_ncols = len(snpinfo_df.columns)
snpinfo_nrows = len(snpinfo_df)
print('No. of rows in', infile0, ':', snpinfo_nrows
, '\nNo. of cols in', infile0, ':', snpinfo_ncols)
afor_df = pd.read_csv(infile1, sep = ',') afor_df = pd.read_csv(infile1, sep = ',')
afor_df_ncols = len(afor_df.columns) afor_ncols = len(afor_df.columns)
afor_df_nrows = len(afor_df) afor_nrows = len(afor_df)
print('No. of rows in', infile1, ':', afor_df_nrows print('No. of rows in', infile1, ':', afor_nrows
, '\nNo. of cols in', infile1, ':', afor_df_ncols) , '\nNo. of cols in', infile1, ':', afor_ncols)
afor_kin_df = pd.read_csv(infile2, sep = ',') afor_kin_df = pd.read_csv(infile2, sep = ',')
afor_kin_df_nrows = len(afor_kin_df) afor_kin_nrows = len(afor_kin_df)
afor_kin_df_ncols = len(afor_kin_df.columns) afor_kin_ncols = len(afor_kin_df.columns)
print('No. of rows in', infile2, ':', afor_kin_df_nrows print('No. of rows in', infile2, ':', afor_kin_nrows
, '\nNo. of cols in', infile2, ':', afor_kin_df_ncols) , '\nNo. of cols in', infile2, ':', afor_kin_ncols)
#======= #%% Process afor_df
# Iterate through the dict, create a lookup dict i.e #1) pull all snp_info so you have ref_allele, etc
# lookup_dict = {three_letter_code: one_letter_code}. # i.e merge afor_df and snpinfo_df
# lookup dict should be the key and the value (you want to create a column for) # find merging column
# Then use this to perform the mapping separetly for wild type and mutant cols.
# The three letter code is extracted using a string match match from the dataframe and then converted
# to 'pandas series'since map only works in pandas series
#=======
gene_regex = gene_match.lower()+'(\w{3})'
print('gene regex being used:', gene_regex)
# initialise a sub dict that is lookup dict for three letter code to 1-letter code left_df = afor_df.copy()
# adding three more cols left_df_nrows = len(left_df)
lookup_dict = dict() left_df_ncols = len(left_df.columns)
for k, v in my_aa_dict.items():
lookup_dict[k] = v['one_letter_code']
# wt = gene_LF1['mutation'].str.extract('gene_p.(\w{3})').squeeze() # converts to a series that map works on
wt = afor_df['mutation'].str.extract(gene_regex).squeeze()
afor_df['wild_type'] = wt.map(lookup_dict)
mut = afor_df['mutation'].str.extract('\d+(\w{3})$').squeeze()
afor_df['mutant_type'] = mut.map(lookup_dict)
# extract position info from mutation column separetly using string match right_df = snpinfo_df.copy()
afor_df['position'] = afor_df['mutation'].str.extract(r'(\d+)') right_df_nrows = len(right_df)
right_df_ncols = len(right_df.columns)
# combine the wild_type+poistion+mutant_type columns to generate common_cols = np.intersect1d(left_df.columns, right_df.columns).tolist()
# mutationinformation (matches mCSM output field) print('Length of common cols:', len(common_cols)
# Remember to use .map(str) for int col types to allow string concatenation , '\ncommon column/s:', common_cols, 'type:', type(common_cols))
afor_df['mutationinformation'] = afor_df['wild_type'] + afor_df['position'].map(str) + afor_df['mutant_type'] print('selecting consistent dtypes for merging (object i.e string)')
print('Created column: mutationinformation' #https://stackoverflow.com/questions/44639772/python-pandas-column-dtype-object-causing-merge-to-fail-with-dtypewarning-colu
, '\n=====================================================================' merging_cols = left_df[common_cols].select_dtypes(include = [object]).columns.tolist()
, afor_df['mutationinformation'].head(10)) print(merging_cols)
nmerging_cols = len(merging_cols)
print(' length of merging cols:', nmerging_cols
, '\nmerging cols:', merging_cols, 'type:', type(merging_cols))
# sanity check # drop duplicates else the expected rows don't match
ncols_add = 4 # beware of hardcoding (3 cols for mcsm style mut + 1 for concatenating them all) print('Checking for duplicates in common col:', common_cols
if len(afor_df.columns) == afor_df_ncols + ncols_add: , '\nNo of duplicates:'
afor_df_ncols = len(afor_df.columns) # update afor_df_ncols after adding cols , len(right_df[right_df.duplicated(common_cols)])
print('PASS: successfully added', ncols_add, 'cols' , '\noriginal length:', right_df_nrows)
, '\nold length:', afor_df_ncols
, '\nnew length:', len(afor_df.columns))
else:
print('FAIL: failed to add cols:'
, '\nExpected cols:', afor_df_ncols + ncols_add
, '\nGot:', len(afor_df.columns))
sys.exit()
#%% Detect mutation format to see if you apply this func
# FIXME
#afor_df.iloc[[0]].str.match('pnca_')
#afor_df.dtypes
#foo = afor_df.loc[:, afor_df.dtypes == object] right_df = right_df[~right_df.duplicated(common_cols)]
right_df_nrows = len(right_df)
genomic_mut_regex = gene_match.lower()+'\w{3}\d+\w{3}' print('\nrevised length:', right_df_nrows)
print('gene regex being used:', genomic_mut_regex)
afor_df[(afor_df == genomic_mut_regex).any(axis = 1)]
#%% Finding common col to merge on
# Define merging column: multiple cols have been used for merge else the common cols
# get suffixes '_x' and '_y' attached
# also, couldn't include 'position' in merging_cols since data types don't match
merging_cols = ['wild_type', 'mutant_type', 'mutationinformation']
ncommon_cols= len(merging_cols)
# checking cross-over of mutations in the two dfs to merge # checking cross-over of mutations in the two dfs to merge
ndiff1 = afor_kin_df_nrows - afor_df['mutationinformation'].isin(afor_kin_df['mutationinformation']).sum() ndiff1 = afor_nrows - afor_df['mutation'].isin(snpinfo_df['mutation']).sum()
print(ndiff1) print('There are', ndiff1, 'mutations with OR, but no snp_info'
ndiff2 = afor_kin_df_nrows - afor_kin_df['mutationinformation'].isin(afor_df['mutationinformation']).sum() , '\nExtracting and writing out file')
print(ndiff2)
#%% combining dfs #afor_df[afor_df['mutation'].isin(snpinfo_df['mutation'])]
missing_mutinfo = afor_df[~afor_df['mutation'].isin(snpinfo_df['mutation'])]
#len(missing_mutinfo.duplicated(common_cols))
#missing_mutinfo.to_csv('infoless_muts.csv')
ndiff2 = snpinfo_nrows - snpinfo_df['mutation'].isin(afor_df['mutation']).sum()
print('There are', ndiff2, 'mutations that do not have OR, but have snp_info')
# Define join type # Define join type
#my_join = 'inner' #my_join = 'inner'
#my_join = 'outer'
#my_join = 'right' #my_join = 'right'
#my_join = 'left' my_join = 'left'
my_join = 'outer'
fail = False print('combing with join:', my_join)
# sanity check: how many muts from afor_kin_df are in afor_df. should be a complete subset combined_df1 = pd.merge(left_df, right_df, on = merging_cols, how = my_join)
if ndiff2 == 0: print('nrows:', len(combined_df1)
print('PASS: all muts in afor_kin_df are present in afor_df' , '\nshape:', combined_df1.shape)
, '\nProceeding with combining the dfs...')
# inner = 252
left_df_nrows - ndiff1
# outer = 331
right_df_nrows + ndiff1
# right = 290
right_df_nrows
# left = 293
left_df_nrows
#%%
# see if you want an extra clause here!
# Define join type
#my_join = 'inner'
#my_join = 'outer'
#my_join = 'right'
my_join = 'left'
fail = False
print('combing with:', my_join)
combined_df1 = pd.merge(left_df, right_df, on = merging_cols, how = my_join)
if my_join == 'inner':
#expected_rows = left_df_nrows - ndiff1
expected_rows = left_df.shape[0] - ndiff1
combined_df = pd.merge(afor_df, afor_kin_df, on = merging_cols, how = my_join) if my_join == 'outer':
#expected_rows = right_df_nrows + ndiff1
expected_rows = right_df.shape[0] + ndiff1
if my_join == ('outer' or 'left') : if my_join == 'right':
print('combing with:', my_join) #expected_rows = right_df_nrows
expected_rows = afor_df_nrows + ndiff1 expected_rows = right_df.shape[0]
if my_join == ('inner' or 'right'): if my_join == 'left':
print('combing with:', my_join) #expected_rows = left_df_nrows
expected_rows = afor_kin_df_nrows expected_rows = left_df.shape[0]
expected_cols = afor_df_ncols + afor_kin_df_ncols - ncommon_cols expected_cols = left_df.shape[1] + right_df.shape[1] - nmerging_cols
if len(combined_df) == expected_rows and len(combined_df.columns) == expected_cols: if len(combined_df1) == expected_rows and len(combined_df1.columns) == expected_cols:
print('PASS: successfully combined dfs with:', my_join, 'join') print('PASS: successfully combined dfs with:', my_join, 'join')
else:
print('FAIL: combined_df\'s expected rows and cols not matched')
fail = True # BAD practice! just a placeholder to avoid code duplication
print('\nExpected no. of rows:', expected_rows
, '\nGot:', len(combined_df)
, '\nExpected no. of cols:', expected_cols
, '\nGot:', len(combined_df.columns))
if fail:
sys.exit('ERROR: combined_df may be incorrectly combined')
else: else:
print('FAIL: numbers mismatch, mutations present in afor_kin_df but not in afor_df') print('FAIL: combined_df\'s expected rows and cols not matched')
sys.exit('ERROR: Not all mutations in the kinship_df are present in the df with other ORs') fail = True
print('\nExpected no. of rows:', expected_rows
, '\nGot:', len(combined_df1)
, '\nExpected no. of cols:', expected_cols
, '\nGot:', len(combined_df1.columns))
if fail:
sys.exit()
#%% check duplicate cols: ones containing suffix '_x' or '_y' # update nrows and ncols
# should only be position afor_info_nrows = len(afor_info_df)
foo = combined_df.filter(regex = r'.*_x|_y', axis = 1) afor_info_ncols = len(afor_info_df.columns)
print(foo.columns) # should only be position #%%
# drop position col containing suffix '_y' and then rename col without suffix
combined_or_df = combined_df.drop(combined_df.filter(regex = r'.*_y').columns, axis = 1)
combined_or_df['position_x'].head()
# renaming columns
combined_or_df.rename(columns = {'position_x': 'position'}, inplace = True)
combined_or_df['position'].head()
# recheck
foo = combined_or_df.filter(regex = r'.*_x|_y', axis = 1)
print(foo.columns) # should only be empty
#%% rearraging columns
print('Dim of df prefromatting:', combined_or_df.shape)
print(combined_or_df.columns)
#%% reorder columns
#https://stackoverflow.com/questions/13148429/how-to-change-the-order-of-dataframe-columns
# setting column's order
output_df = combined_or_df[['mutation',
'mutationinformation',
'wild_type',
'position',
'mutant_type',
'chr_num_allele',
'ref_allele',
'alt_allele',
'mut_info',
'mut_type',
'gene_id',
'gene_number',
'mut_region',
'reference_allele',
'alternate_allele',
'chromosome_number',
'af',
'af_kin',
'or_kin',
'or_logistic',
'or_mychisq',
'est_chisq',
'or_fisher',
'ci_low_logistic',
'ci_hi_logistic',
'ci_low_fisher',
'ci_hi_fisher',
'pwald_kin',
'pval_logistic',
'pval_fisher',
'pval_chisq',
'beta_logistic',
'beta_kin',
'se_logistic',
'se_kin',
'zval_logistic',
'logl_H1_kin',
'l_remle_kin',
'n_diff',
'tot_diff',
'n_miss']]
# sanity check after rearranging
if combined_or_df.shape == output_df.shape and set(combined_or_df.columns) == set(output_df.columns):
print('PASS: Successfully formatted df with rearranged columns')
else:
sys.exit('FAIL: something went wrong when rearranging columns!')
#%% write file
print('\n====================================================================='
, '\nWriting output file:\n', outfile
, '\nNo.of rows:', len(output_df)
, '\nNo. of cols:', len(output_df.columns))
output_df.to_csv(outfile, index = False)