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

ran struc param analysis

Browse source
This commit is contained in:
Tanushree Tunstall 2020-06-17 19:36:02 +01:00
parent 96da4d8ed5
commit 2cebd338ba
5 changed files with 373 additions and 382 deletions
Download patch file Download diff file Expand all files Collapse all files

335
scripts/combine_dfs.py
View file

@ -46,10 +46,8 @@ os.getcwd()
#======================================================================= #=======================================================================
#%% 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 = None)
arg_parser.add_argument('-g', '--gene', help='gene name', default = 'pncA') # case sensitive arg_parser.add_argument('-g', '--gene', help='gene name', default = None) # case sensitive
#arg_parser.add_argument('-d', '--drug', help='drug name', default = 'TESTDRUG')
#arg_parser.add_argument('-g', '--gene', help='gene name (case sensitive)', default = 'testGene') # case sensitive
args = arg_parser.parse_args() args = arg_parser.parse_args()
#======================================================================= #=======================================================================
#%% variable assignment: input and output #%% variable assignment: input and output
@ -101,178 +99,178 @@ print('Output filename:', out_filename
#%% function/methd to combine 4 dfs #%% function/methd to combine 4 dfs
def combine_dfs(dssp_csv, kd_csv, rd_csv, mcsm_csv, out_combined_csv): def combine_dfs(dssp_csv, kd_csv, rd_csv, mcsm_csv, out_combined_csv):
""" """
Combine 4 dfs Combine 4 dfs
@param dssp_df: csv file (output from dssp_df.py) @param dssp_df: csv file (output from dssp_df.py)
@type dssp_df: string @type dssp_df: string
@param kd_df: csv file (output from kd_df.py) @param kd_df: csv file (output from kd_df.py)
@type ks_df: string @type ks_df: string
@param rd_df: csv file (output from rd_df.py) @param rd_df: csv file (output from rd_df.py)
@type rd_df: string @type rd_df: string
# FIXME # FIXME
@param mcsm_df: csv file (output of mcsm pipeline)CHECK} @param mcsm_df: csv file (output of mcsm pipeline)CHECK}
@type mcsm_df: string @type mcsm_df: string
@param out_combined_csv: csv file output @param out_combined_csv: csv file output
@type out_combined_csv: string @type out_combined_csv: string
@return: none, writes combined df as csv @return: none, writes combined df as csv
""" """
#======================== #========================
# read input csv files to combine # read input csv files to combine
#======================== #========================
dssp_df = pd.read_csv(dssp_csv, sep = ',') dssp_df = pd.read_csv(dssp_csv, sep = ',')
kd_df = pd.read_csv(kd_csv, sep = ',') kd_df = pd.read_csv(kd_csv, sep = ',')
rd_df = pd.read_csv(rd_csv, sep = ',') rd_df = pd.read_csv(rd_csv, sep = ',')
mcsm_df = pd.read_csv(mcsm_csv, sep = ',') mcsm_df = pd.read_csv(mcsm_csv, sep = ',')
print('Reading input files:' print('Reading input files:'
, '\ndssp file:', dssp_csv , '\ndssp file:', dssp_csv
, '\nNo. of rows:', len(dssp_df) , '\nNo. of rows:', len(dssp_df)
, '\nNo. of cols:', len(dssp_df.columns) , '\nNo. of cols:', len(dssp_df.columns)
, '\nColumn names:', dssp_df.columns , '\nColumn names:', dssp_df.columns
, '\n===================================================================' , '\n==================================================================='
, '\nkd file:', kd_csv , '\nkd file:', kd_csv
, '\nNo. of rows:', len(kd_df) , '\nNo. of rows:', len(kd_df)
, '\nNo. of cols:', len(kd_df.columns) , '\nNo. of cols:', len(kd_df.columns)
, '\nColumn names:', kd_df.columns , '\nColumn names:', kd_df.columns
, '\n===================================================================' , '\n==================================================================='
, '\nrd file:', rd_csv , '\nrd file:', rd_csv
, '\nNo. of rows:', len(rd_df) , '\nNo. of rows:', len(rd_df)
, '\nNo. of cols:', len(rd_df.columns) , '\nNo. of cols:', len(rd_df.columns)
, '\nColumn names:', rd_df.columns , '\nColumn names:', rd_df.columns
, '\n===================================================================' , '\n==================================================================='
, '\nrd file:', mcsm_csv , '\nrd file:', mcsm_csv
, '\nNo. of rows:', len(mcsm_df) , '\nNo. of rows:', len(mcsm_df)
, '\nNo. of cols:', len(mcsm_df.columns) , '\nNo. of cols:', len(mcsm_df.columns)
, '\nColumn names:', mcsm_df.columns , '\nColumn names:', mcsm_df.columns
, '\n===================================================================') , '\n===================================================================')
#======================== #========================
# merge 1 (combined_df1) # merge 1 (combined_df1)
# concatenating 3dfs: # concatenating 3dfs:
# dssp_df, kd_df, rd_df # dssp_df, kd_df, rd_df
#======================== #========================
print('starting first merge...\n') print('starting first merge...\n')
# checking no. of rows # checking no. of rows
print('Checking if no. of rows of the 3 dfs are equal:\n' print('Checking if no. of rows of the 3 dfs are equal:\n'
, len(dssp_df) == len(kd_df) == len(rd_df) , len(dssp_df) == len(kd_df) == len(rd_df)
, '\nReason: fasta files and pdb files vary since not all pos are part of the structure' , '\nReason: fasta files and pdb files vary since not all pos are part of the structure'
, '\n===================================================================') , '\n===================================================================')
# variables for sanity checks # variables for sanity checks
expected_rows_df1 = max(len(dssp_df), len(kd_df), len(rd_df)) expected_rows_df1 = max(len(dssp_df), len(kd_df), len(rd_df))
# beware of harcoding! used for sanity check # beware of harcoding! used for sanity check
ndfs = 3 ndfs = 3
ncol_merge = 1 ncol_merge = 1
offset = ndfs- ncol_merge offset = ndfs- ncol_merge
expected_cols_df1 = len(dssp_df.columns) + len(kd_df.columns) + len(rd_df.columns) - offset expected_cols_df1 = len(dssp_df.columns) + len(kd_df.columns) + len(rd_df.columns) - offset
print('Merge 1:' print('Merge 1:'
, '\ncombining 3dfs by commom col: position' , '\ncombining 3dfs by commom col: position'
, '\nExpected nrows in combined_df:', expected_rows_df1 , '\nExpected nrows in combined_df:', expected_rows_df1
, '\nExpected ncols in combined_df:', expected_cols_df1 , '\nExpected ncols in combined_df:', expected_cols_df1
, '\nResetting the common col as the index' , '\nResetting the common col as the index'
, '\n===================================================================') , '\n===================================================================')
#dssp_df.set_index('position', inplace = True) #dssp_df.set_index('position', inplace = True)
#kd_df.set_index('position', inplace = True) #kd_df.set_index('position', inplace = True)
#rd_df.set_index('position', inplace =True) #rd_df.set_index('position', inplace =True)
#combined_df = pd.concat([dssp_df, kd_df, rd_df], axis = 1, sort = False).reset_index() #combined_df = pd.concat([dssp_df, kd_df, rd_df], axis = 1, sort = False).reset_index()
#combined_df.rename(columns = {'index':'position'}) #combined_df.rename(columns = {'index':'position'})
combined_df1 = pd.concat( combined_df1 = pd.concat(
(my_index.set_index('position') for my_index in [dssp_df, kd_df, rd_df]) (my_index.set_index('position') for my_index in [dssp_df, kd_df, rd_df])
, axis = 1, join = 'outer').reset_index() , axis = 1, join = 'outer').reset_index()
# sanity check # sanity check
print('Checking dimensions of concatenated df1...') print('Checking dimensions of concatenated df1...')
if len(combined_df1) == expected_rows_df1 and len(combined_df1.columns) == expected_cols_df1: if len(combined_df1) == expected_rows_df1 and len(combined_df1.columns) == expected_cols_df1:
print('PASS: combined df has expected dimensions' print('PASS: combined df has expected dimensions'
, '\nNo. of rows in combined df:', len(combined_df1) , '\nNo. of rows in combined df:', len(combined_df1)
, '\nNo. of cols in combined df:', len(combined_df1.columns) , '\nNo. of cols in combined df:', len(combined_df1.columns)
, '\n===============================================================') , '\n===============================================================')
else: else:
print('FAIL: combined df does not have expected dimensions' print('FAIL: combined df does not have expected dimensions'
, '\nNo. of rows in combined df:', len(combined_df1) , '\nNo. of rows in combined df:', len(combined_df1)
, '\nNo. of cols in combined df:', len(combined_df1.columns) , '\nNo. of cols in combined df:', len(combined_df1.columns)
, '\n===============================================================') , '\n===============================================================')
#======================== #========================
# merge 2 (combined_df2) # merge 2 (combined_df2)
# concatenating 2dfs: # concatenating 2dfs:
# mcsm_df, combined_df1 (result of merge1) # mcsm_df, combined_df1 (result of merge1)
# sort the cols # sort the cols
#======================== #========================
print('starting second merge...\n') print('starting second merge...\n')
# rename col 'Position' in mcsm_df to lowercase 'position' # rename col 'Position' in mcsm_df to lowercase 'position'
# as it matches the combined_df1 colname to perfom merge # as it matches the combined_df1 colname to perfom merge
#mcsm_df.columns #mcsm_df.columns
#mcsm_df.rename(columns = {'Position':'position'}) # not working! #mcsm_df.rename(columns = {'Position':'position'}) # not working!
# copy 'Position' column with the correct colname # copy 'Position' column with the correct colname
print('Firstly, copying \'Position\' col and renaming \'position\' to allow merging' print('Firstly, copying \'Position\' col and renaming \'position\' to allow merging'
, '\nNo. of cols before copying: ', len(mcsm_df.columns)) , '\nNo. of cols before copying: ', len(mcsm_df.columns))
mcsm_df['position'] = mcsm_df['Position'] mcsm_df['position'] = mcsm_df['Position']
print('No. of cols after copying: ', len(mcsm_df.columns)) print('No. of cols after copying: ', len(mcsm_df.columns))
# sanity check # sanity check
if mcsm_df['position'].equals(mcsm_df['Position']): if mcsm_df['position'].equals(mcsm_df['Position']):
print('PASS: Copying worked correctly' print('PASS: Copying worked correctly'
, '\ncopied col matches original column' , '\ncopied col matches original column'
, '\n===============================================================') , '\n===============================================================')
else: else:
print('FAIL: copied col does not match original column' print('FAIL: copied col does not match original column'
, '\n================================================================') , '\n================================================================')
# variables for sanity checks # variables for sanity checks
expected_rows_df2 = len(mcsm_df) expected_rows_df2 = len(mcsm_df)
# beware of harcoding! used for sanity check # beware of harcoding! used for sanity check
ndfs = 2 ndfs = 2
ncol_merge = 1 ncol_merge = 1
offset = ndfs - ncol_merge offset = ndfs - ncol_merge
expected_cols_df2 = len(mcsm_df.columns) + len(combined_df1.columns) - offset expected_cols_df2 = len(mcsm_df.columns) + len(combined_df1.columns) - offset
print('Merge 2:' print('Merge 2:'
, '\ncombining 2dfs by commom col: position' , '\ncombining 2dfs by commom col: position'
, '\nExpected nrows in combined_df:', expected_rows_df2 , '\nExpected nrows in combined_df:', expected_rows_df2
, '\nExpected ncols in combined_df:', expected_cols_df2 , '\nExpected ncols in combined_df:', expected_cols_df2
, '\n===================================================================') , '\n===================================================================')
combined_df2 = mcsm_df.merge(combined_df1, on = 'position') combined_df2 = mcsm_df.merge(combined_df1, on = 'position')
# sanity check # sanity check
print('Checking dimensions of concatenated df2...') print('Checking dimensions of concatenated df2...')
if len(combined_df2) == expected_rows_df2 and len(combined_df2.columns) == expected_cols_df2: if len(combined_df2) == expected_rows_df2 and len(combined_df2.columns) == expected_cols_df2:
print('PASS: combined df2 has expected dimensions' print('PASS: combined df2 has expected dimensions'
, '\nNo. of rows in combined df:', len(combined_df2) , '\nNo. of rows in combined df:', len(combined_df2)
, '\nNo. of cols in combined df:', len(combined_df2.columns) , '\nNo. of cols in combined df:', len(combined_df2.columns)
, '\n===============================================================') , '\n===============================================================')
else: else:
print('FAIL: combined df2 does not have expected dimensions' print('FAIL: combined df2 does not have expected dimensions'
, '\nNo. of rows in combined df:', len(combined_df2) , '\nNo. of rows in combined df:', len(combined_df2)
, '\nNo. of cols in combined df:', len(combined_df2.columns) , '\nNo. of cols in combined df:', len(combined_df2.columns)
, '\n===============================================================') , '\n===============================================================')
#=============== #===============
# writing file # writing file
#=============== #===============
print('Writing file:' print('Writing file:'
, '\nFilename:', out_combined_csv , '\nFilename:', out_combined_csv
# , '\nPath:', outdir # , '\nPath:', outdir
, '\nExpected no. of rows:', len(combined_df2) , '\nExpected no. of rows:', len(combined_df2)
, '\nExpected no. of cols:', len(combined_df2.columns) , '\nExpected no. of cols:', len(combined_df2.columns)
, '\n=========================================================') , '\n=========================================================')
combined_df2.to_csv(out_combined_csv, header = True, index = False) combined_df2.to_csv(out_combined_csv, header = True, index = False)
#%% end of function #%% end of function
#======================================================================= #=======================================================================
@ -280,19 +278,18 @@ def combine_dfs(dssp_csv, kd_csv, rd_csv, mcsm_csv, out_combined_csv):
#combine_dfs(infile1, infile2, infile3, infile4, outfile) #combine_dfs(infile1, infile2, infile3, infile4, outfile)
#======================================================================= #=======================================================================
def main(): def main():
print('Combining 4 dfs:\n' print('Combining 4 dfs:\n'
, in_filename1, '\n' , in_filename1, '\n'
, in_filename2, '\n' , in_filename2, '\n'
, in_filename3, '\n' , in_filename3, '\n'
, in_filename4, '\n' , in_filename4, '\n'
, 'output csv:', out_filename) , 'output csv:', out_filename)
combine_dfs(infile1, infile2, infile3, infile4, outfile) combine_dfs(infile1, infile2, infile3, infile4, outfile)
print('Finished Writing file:' print('Finished Writing file:'
, '\nFilename:', out_filename , '\nFilename:', outfile
, '\nPath:', outdir ## , '\nNo. of rows:', ''
## , '\nNo. of rows:', '' ## , '\nNo. of cols:', ''
## , '\nNo. of cols:', '' , '\n===========================================================')
, '\n===========================================================')
if __name__ == '__main__': if __name__ == '__main__':
main() main()

30
scripts/data_extraction.py
View file

@ -57,8 +57,8 @@ args = arg_parser.parse_args()
drug = args.drug drug = args.drug
gene = args.gene gene = args.gene
gene_match = gene + '_p.' gene_match = gene + '_p.'
# building cols to extract # building cols to extract
dr_muts_col = 'dr_mutations_' + drug dr_muts_col = 'dr_mutations_' + drug
other_muts_col = 'other_mutations_' + drug other_muts_col = 'other_mutations_' + drug
@ -80,8 +80,8 @@ datadir = homedir + '/' + 'git/Data'
#======= #=======
# input # input
#======= #=======
#in_filename = 'original_tanushree_data_v2.csv' in_filename = 'original_tanushree_data_v2.csv'
in_filename = 'mtb_gwas_v3.csv' #in_filename = 'mtb_gwas_v3.csv'
infile = datadir + '/' + in_filename infile = datadir + '/' + in_filename
print('Input file: ', infile print('Input file: ', infile
, '\n============================================================') , '\n============================================================')
@ -1028,25 +1028,25 @@ del(k, v, wt, mut, lookup_dict)
######## ########
# combine the wild_type+poistion+mutant_type columns to generate # combine the wild_type+poistion+mutant_type columns to generate
# Mutationinformation (matches mCSM output field) # mutationinformation (matches mCSM output field)
# Remember to use .map(str) for int col types to allow string concatenation # Remember to use .map(str) for int col types to allow string concatenation
######### #########
gene_LF1['Mutationinformation'] = gene_LF1['wild_type'] + gene_LF1.position.map(str) + gene_LF1['mutant_type'] gene_LF1['mutationinformation'] = gene_LF1['wild_type'] + gene_LF1.position.map(str) + gene_LF1['mutant_type']
print('Created column: Mutationinformation' print('Created column: mutationinformation'
, '\n=====================================================================' , '\n====================================================================='
, gene_LF1.Mutationinformation.head(10)) , gene_LF1.mutationinformation.head(10))
#%% Write file: mCSM muts #%% Write file: mCSM muts
snps_only = pd.DataFrame(gene_LF1['Mutationinformation'].unique()) snps_only = pd.DataFrame(gene_LF1['mutationinformation'].unique())
snps_only.head() snps_only.head()
# assign column name # assign column name
snps_only.columns = ['Mutationinformation'] snps_only.columns = ['mutationinformation']
# count how many positions this corresponds to # count how many positions this corresponds to
pos_only = pd.DataFrame(gene_LF1['position'].unique()) pos_only = pd.DataFrame(gene_LF1['position'].unique())
print('Checking NA in snps...')# should be 0 print('Checking NA in snps...')# should be 0
if snps_only.Mutationinformation.isna().sum() == 0: if snps_only.mutationinformation.isna().sum() == 0:
print ('PASS: NO NAs/missing entries for SNPs' print ('PASS: NO NAs/missing entries for SNPs'
, '\n===============================================================') , '\n===============================================================')
else: else:
@ -1090,27 +1090,27 @@ print('Finished writing:', out_filename3
del(out_filename3) del(out_filename3)
#%% write file: mCSM style but with repitions for MSA and logo plots #%% write file: mCSM style but with repitions for MSA and logo plots
all_muts_msa = pd.DataFrame(gene_LF1['Mutationinformation']) all_muts_msa = pd.DataFrame(gene_LF1['mutationinformation'])
all_muts_msa.head() all_muts_msa.head()
# assign column name # assign column name
all_muts_msa.columns = ['Mutationinformation'] all_muts_msa.columns = ['mutationinformation']
# make sure it is string # make sure it is string
all_muts_msa.columns.dtype all_muts_msa.columns.dtype
# sort the column # sort the column
all_muts_msa_sorted = all_muts_msa.sort_values(by = 'Mutationinformation') all_muts_msa_sorted = all_muts_msa.sort_values(by = 'mutationinformation')
# create an extra column with protein name # create an extra column with protein name
all_muts_msa_sorted = all_muts_msa_sorted.assign(fasta_name = '3PL1') all_muts_msa_sorted = all_muts_msa_sorted.assign(fasta_name = '3PL1')
all_muts_msa_sorted.head() all_muts_msa_sorted.head()
# rearrange columns so the fasta name is the first column (required for mutate.script) # rearrange columns so the fasta name is the first column (required for mutate.script)
all_muts_msa_sorted = all_muts_msa_sorted[['fasta_name', 'Mutationinformation']] all_muts_msa_sorted = all_muts_msa_sorted[['fasta_name', 'mutationinformation']]
all_muts_msa_sorted.head() all_muts_msa_sorted.head()
print('Checking NA in snps...')# should be 0 print('Checking NA in snps...')# should be 0
if all_muts_msa.Mutationinformation.isna().sum() == 0: if all_muts_msa.mutationinformation.isna().sum() == 0:
print ('PASS: NO NAs/missing entries for SNPs' print ('PASS: NO NAs/missing entries for SNPs'
, '\n===============================================================') , '\n===============================================================')
else: else:

10
scripts/dssp_df.py
View file

@ -30,10 +30,8 @@ os.getcwd()
#======================================================================= #=======================================================================
#%% 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 = None)
#arg_parser.add_argument('-g', '--gene', help='gene name', default = 'pncA') # case sensitive arg_parser.add_argument('-g', '--gene', help='gene name (case sensitive)', default = None) # case sensitive
arg_parser.add_argument('-d', '--drug', help='drug name', default = 'TESTDRUG')
arg_parser.add_argument('-g', '--gene', help='gene name (case sensitive)', default = 'testGene') # case sensitive
args = arg_parser.parse_args() args = arg_parser.parse_args()
#======================================================================= #=======================================================================
#%% variable assignment: input and output #%% variable assignment: input and output
@ -49,6 +47,8 @@ args = arg_parser.parse_args()
drug = args.drug drug = args.drug
gene = args.gene gene = args.gene
gene_match = gene + '_p.'
#========== #==========
# data dir # data dir
#========== #==========
@ -147,7 +147,7 @@ def extract_chain_dssp(inputpdbfile):
return pdbchainlist return pdbchainlist
#======================================================================= #=======================================================================
#%% write csv of processed dssp output #%% write csv of processed dssp output
def dssp_to_csv(inputdsspfile, outfile, pdbchainlist): def dssp_to_csv(inputdsspfile, outfile, pdbchainlist = ['A']):
""" """
Create a df from a dssp file containing ASA, RSA, SS for all chains Create a df from a dssp file containing ASA, RSA, SS for all chains

233
scripts/kd_df.py
View file

@ -39,10 +39,8 @@ os.getcwd()
#======================================================================= #=======================================================================
#%% 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 = None)
#arg_parser.add_argument('-g', '--gene', help='gene name', default = 'pncA') # case sensitive arg_parser.add_argument('-g', '--gene', help='gene name', default = None)
arg_parser.add_argument('-d', '--drug', help='drug name', default = 'DRUGNAME')
arg_parser.add_argument('-g', '--gene', help='gene name', default = 'geneName')
#arg_parser.add_argument('-p', '--plot', help='show plot', action='store_true') #arg_parser.add_argument('-p', '--plot', help='show plot', action='store_true')
args = arg_parser.parse_args() args = arg_parser.parse_args()
#======================================================================= #=======================================================================
@ -81,8 +79,8 @@ print('Output filename:', out_filename
#%% end of variable assignment for input and output files #%% end of variable assignment for input and output files
#======================================================================= #=======================================================================
#%% kd values from fasta file and output csv #%% kd values from fasta file and output csv
def kd_to_csv(inputfasta, outputkdcsv, windowsize): def kd_to_csv(inputfasta, outputkdcsv, windowsize = 3):
""" """
Calculate kd (hydropathy values) from input fasta file Calculate kd (hydropathy values) from input fasta file
@param inputfasta: fasta file @param inputfasta: fasta file
@ -96,138 +94,137 @@ def kd_to_csv(inputfasta, outputkdcsv, windowsize):
@return: none, writes kd values df as csv @return: none, writes kd values df as csv
""" """
#======================== #========================
# read input fasta file # read input fasta file
#======================== #========================
fh = open(inputfasta) fh = open(inputfasta)
for record in SeqIO.parse(fh, 'fasta'): for record in SeqIO.parse(fh, 'fasta'):
id = record.id id = record.id
seq = record.seq seq = record.seq
num_residues = len(seq) num_residues = len(seq)
fh.close() fh.close()
sequence = str(seq) sequence = str(seq)
X = ProteinAnalysis(sequence) X = ProteinAnalysis(sequence)
#=================== #===================
# calculate KD values: same as the expasy server # calculate KD values: same as the expasy server
#=================== #===================
my_window = windowsize my_window = windowsize
offset = round((my_window/2)-0.5) offset = round((my_window/2)-0.5)
# edge weight is set to default (100%) # edge weight is set to default (100%)
kd_values = (X.protein_scale(ProtParamData.kd , window = my_window)) kd_values = (X.protein_scale(ProtParamData.kd , window = my_window))
# sanity checks # sanity checks
print('Sequence Length:', num_residues) print('Sequence Length:', num_residues)
print('kd_values Length:',len(kd_values)) print('kd_values Length:',len(kd_values))
print('Window Length:', my_window) print('Window Length:', my_window)
print('Window Offset:', offset) print('Window Offset:', offset)
print('=================================================================') print('=================================================================')
print('Checking:len(kd values) is as expected for the given window size & offset...') print('Checking:len(kd values) is as expected for the given window size & offset...')
expected_length = num_residues - (my_window - offset) expected_length = num_residues - (my_window - offset)
if len(kd_values) == expected_length: if len(kd_values) == expected_length:
print('PASS: expected and actual length of kd values match') print('PASS: expected and actual length of kd values match')
else: else:
print('FAIL: length mismatch' print('FAIL: length mismatch'
,'\nExpected length:', expected_length ,'\nExpected length:', expected_length
,'\nActual length:', len(kd_values) ,'\nActual length:', len(kd_values)
, '\n=========================================================') , '\n=========================================================')
#=================== #===================
# creating two dfs # creating two dfs
#=================== #===================
# 1) aa sequence and 2) kd_values. Then reset index for each df # 1) aa sequence and 2) kd_values. Then reset index for each df
# which will allow easy merging of the two dfs. # which will allow easy merging of the two dfs.
# df1: df of aa seq with index reset to start from 1 # df1: df of aa seq with index reset to start from 1
# (reflective of the actual aa position in a sequence) # (reflective of the actual aa position in a sequence)
# Name column of wt as 'wild_type' to be the same name used # Name column of wt as 'wild_type' to be the same name used
# in the file required for merging later. # in the file required for merging later.
dfSeq = pd.DataFrame({'wild_type_kd':list(sequence)}) dfSeq = pd.DataFrame({'wild_type_kd':list(sequence)})
dfSeq.index = np.arange(1, len(dfSeq) + 1) # python is not inclusive dfSeq.index = np.arange(1, len(dfSeq) + 1) # python is not inclusive
# df2: df of kd_values with index reset to start from offset + 1 and # df2: df of kd_values with index reset to start from offset + 1 and
# subsequent matched length of the kd_values # subsequent matched length of the kd_values
dfVals = pd.DataFrame({'kd_values':kd_values}) dfVals = pd.DataFrame({'kd_values':kd_values})
dfVals.index = np.arange(offset + 1, len(dfVals) + 1 + offset) dfVals.index = np.arange(offset + 1, len(dfVals) + 1 + offset)
# sanity checks # sanity checks
max(dfVals['kd_values']) max(dfVals['kd_values'])
min(dfVals['kd_values']) min(dfVals['kd_values'])
#=================== #===================
# concatenating dfs # concatenating dfs
#=================== #===================
# Merge the two on index # Merge the two on index
# (as these are now reflective of the aa position numbers): df1 and df2 # (as these are now reflective of the aa position numbers): df1 and df2
# This will introduce NaN where there is missing values. In our case this # This will introduce NaN where there is missing values. In our case this
# will be 2 (first and last ones based on window size and offset) # will be 2 (first and last ones based on window size and offset)
kd_df = pd.concat([dfSeq, dfVals], axis = 1) kd_df = pd.concat([dfSeq, dfVals], axis = 1)
#============================ #============================
# renaming index to position # renaming index to position
#============================ #============================
kd_df = kd_df.rename_axis('position') kd_df = kd_df.rename_axis('position')
kd_df.head kd_df.head
print('Checking: position col i.e. index should be numeric') print('Checking: position col i.e. index should be numeric')
if kd_df.index.dtype == 'int64': if kd_df.index.dtype == 'int64':
print('PASS: position col is numeric' print('PASS: position col is numeric'
, '\ndtype is:', kd_df.index.dtype) , '\ndtype is:', kd_df.index.dtype)
else: else:
print('FAIL: position col is not numeric' print('FAIL: position col is not numeric'
, '\nConverting to numeric') , '\nConverting to numeric')
kd_df.index.astype('int64') kd_df.index.astype('int64')
print('Checking dtype for after conversion:\n' print('Checking dtype for after conversion:\n'
, '\ndtype is:', kd_df.index.dtype , '\ndtype is:', kd_df.index.dtype
, '\n=========================================================') , '\n=========================================================')
#=============== #===============
# writing file # writing file
#=============== #===============
print('Writing file:' print('Writing file:'
, '\nFilename:', outputkdcsv , '\nFilename:', outputkdcsv
# , '\nPath:', outdir # , '\nPath:', outdir
, '\nExpected no. of rows:', len(kd_df) , '\nExpected no. of rows:', len(kd_df)
, '\nExpected no. of cols:', len(kd_df.columns) , '\nExpected no. of cols:', len(kd_df.columns)
, '\n=============================================================') , '\n=============================================================')
kd_df.to_csv(outputkdcsv, header = True, index = True) kd_df.to_csv(outputkdcsv, header = True, index = True)
#=============== #===============
# plot: optional! # plot: optional!
#=============== #===============
# http://www.dalkescientific.com/writings/NBN/plotting.html # http://www.dalkescientific.com/writings/NBN/plotting.html
# FIXME: save fig # FIXME: save fig
# extract just pdb if from 'id' to pass to title of plot # extract just pdb if from 'id' to pass to title of plot
# foo = re.match(r'(^[0-9]{1}\w{3})', id).groups(1) # foo = re.match(r'(^[0-9]{1}\w{3})', id).groups(1)
# if doplot: # if doplot:
plot(kd_values, linewidth = 1.0) plot(kd_values, linewidth = 1.0)
#axis(xmin = 1, xmax = num_residues) #axis(xmin = 1, xmax = num_residues)
xlabel('Residue Number') xlabel('Residue Number')
ylabel('Hydrophobicity') ylabel('Hydrophobicity')
title('K&D Hydrophobicity for ' + id) title('K&D Hydrophobicity for ' + id)
show() show()
#%% end of function #%% end of function
#======================================================================= #=======================================================================
#%% call function #%% call function
#kd_to_csv(infile, outfile, windowsize = 3) #kd_to_csv(infile, outfile, windowsize = 3)
#======================================================================= #=======================================================================
def main(): def main():
print('Running hydropathy calcs with following params\n' print('Running hydropathy calcs with following params\n'
, in_filename , in_filename
, '\noutfile:', out_filename) , '\noutfile:', out_filename)
kd_to_csv(infile, outfile, 3) kd_to_csv(infile, outfile, 3)
print('Finished writing file:' print('Finished writing file:'
, '\nFilename:', out_filename , '\nFilename:', outfile
, '\nPath:', outdir , '\n=============================================================')
, '\n=============================================================')
if __name__ == '__main__': if __name__ == '__main__':
main() main()
#%% end of script #%% end of script
#======================================================================= #=======================================================================

147
scripts/rd_df.py
View file

@ -31,10 +31,8 @@ os.getcwd()
#======================================================================= #=======================================================================
#%% 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 = None)
#arg_parser.add_argument('-g', '--gene', help='gene name', default = 'pncA') # case sensitive arg_parser.add_argument('-g', '--gene', help='gene name', default = None) # case sensitive
arg_parser.add_argument('-d', '--drug', help='drug name', default = 'TESTDRUG')
arg_parser.add_argument('-g', '--gene', help='gene name (case sensitive)', default = 'testGene') # case sensitive
args = arg_parser.parse_args() args = arg_parser.parse_args()
#======================================================================= #=======================================================================
#%% variable assignment: input and output #%% variable assignment: input and output
@ -72,7 +70,7 @@ print('Output filename:', out_filename
#======================================================================= #=======================================================================
#%% rd values from <gene>_rd.tsv values #%% rd values from <gene>_rd.tsv values
def rd_to_csv(inputtsv, outputrdcsv): def rd_to_csv(inputtsv, outputrdcsv):
""" """
Calculate kd (hydropathy values) from input fasta file Calculate kd (hydropathy values) from input fasta file
@param inputtsv: tsv file downloaded from {INSERT LINK} @param inputtsv: tsv file downloaded from {INSERT LINK}
@ -83,76 +81,76 @@ def rd_to_csv(inputtsv, outputrdcsv):
@return: none, writes rd values df as csv @return: none, writes rd values df as csv
""" """
#======================== #========================
# read downloaded tsv file # read downloaded tsv file
#======================== #========================
#%% Read input file #%% Read input file
rd_data = pd.read_csv(inputtsv, sep = '\t') rd_data = pd.read_csv(inputtsv, sep = '\t')
print('Reading input file:', inputtsv print('Reading input file:', inputtsv
, '\nNo. of rows:', len(rd_data) , '\nNo. of rows:', len(rd_data)
, '\nNo. of cols:', len(rd_data.columns)) , '\nNo. of cols:', len(rd_data.columns))
print('Column names:', rd_data.columns print('Column names:', rd_data.columns
, '\n===============================================================') , '\n===============================================================')
#======================== #========================
# creating position col # creating position col
#======================== #========================
# Extracting residue number from index and assigning # Extracting residue number from index and assigning
# the values to a column [position]. Then convert the position col to numeric. # the values to a column [position]. Then convert the position col to numeric.
rd_data['position'] = rd_data.index.str.extract('([0-9]+)').values rd_data['position'] = rd_data.index.str.extract('([0-9]+)').values
# converting position to numeric # converting position to numeric
rd_data['position'] = pd.to_numeric(rd_data['position']) rd_data['position'] = pd.to_numeric(rd_data['position'])
rd_data['position'].dtype rd_data['position'].dtype
print('Extracted residue num from index and assigned as a column:' print('Extracted residue num from index and assigned as a column:'
, '\ncolumn name: position' , '\ncolumn name: position'
, '\ntotal no. of cols now:', len(rd_data.columns) , '\ntotal no. of cols now:', len(rd_data.columns)
, '\n=============================================================') , '\n=============================================================')
#======================== #========================
# Renaming amino-acid # Renaming amino-acid
# and all-atom cols # and all-atom cols
#======================== #========================
print('Renaming columns:' print('Renaming columns:'
, '\ncolname==> # chain:residue: wt_3letter_caps' , '\ncolname==> # chain:residue: wt_3letter_caps'
, '\nYES... the column name *actually* contains a # ..!' , '\nYES... the column name *actually* contains a # ..!'
, '\ncolname==> all-atom: rd_values' , '\ncolname==> all-atom: rd_values'
, '\n=============================================================') , '\n=============================================================')
rd_data.rename(columns = {'# chain:residue':'wt_3letter_caps', 'all-atom':'rd_values'}, inplace = True) rd_data.rename(columns = {'# chain:residue':'wt_3letter_caps', 'all-atom':'rd_values'}, inplace = True)
print('Column names:', rd_data.columns) print('Column names:', rd_data.columns)
#======================== #========================
# extracting df with the # extracting df with the
# desired columns # desired columns
#======================== #========================
print('Extracting relevant columns for writing df as csv') print('Extracting relevant columns for writing df as csv')
rd_df = rd_data[['position','rd_values','wt_3letter_caps']] rd_df = rd_data[['position','rd_values','wt_3letter_caps']]
if len(rd_df) == len(rd_data): if len(rd_df) == len(rd_data):
print('PASS: extracted df has expected no. of rows' print('PASS: extracted df has expected no. of rows'
,'\nExtracted df dim:' ,'\nExtracted df dim:'
,'\nNo. of rows:', len(rd_df) ,'\nNo. of rows:', len(rd_df)
,'\nNo. of cols:', len(rd_df.columns)) ,'\nNo. of cols:', len(rd_df.columns))
else: else:
print('FAIL: no. of rows mimatch' print('FAIL: no. of rows mimatch'
, '\nExpected no. of rows:', len(rd_data) , '\nExpected no. of rows:', len(rd_data)
, '\nGot no. of rows:', len(rd_df) , '\nGot no. of rows:', len(rd_df)
, '\n=====================================================') , '\n=====================================================')
#=============== #===============
# writing file # writing file
#=============== #===============
print('Writing file:' print('Writing file:'
, '\nFilename:', outputrdcsv , '\nFilename:', outputrdcsv
# , '\nPath:', outdir # , '\nPath:', outdir
# , '\nExpected no. of rows:', len(rd_df) # , '\nExpected no. of rows:', len(rd_df)
# , '\nExpected no. of cols:', len(rd_df.columns) # , '\nExpected no. of cols:', len(rd_df.columns)
, '\n=========================================================') , '\n=========================================================')
rd_df.to_csv(outputrdcsv, header = True, index = False) rd_df.to_csv(outputrdcsv, header = True, index = False)
#%% end of function #%% end of function
#======================================================================= #=======================================================================
@ -160,16 +158,15 @@ def rd_to_csv(inputtsv, outputrdcsv):
#rd_to_csv(infile, outfile) #rd_to_csv(infile, outfile)
#======================================================================= #=======================================================================
def main(): def main():
print('residue depth using the following params\n' print('residue depth using the following params\n'
, in_filename , in_filename
, '\noutfile:', out_filename) , '\noutfile:', out_filename)
rd_to_csv(infile, outfile) rd_to_csv(infile, outfile)
print('Finished Writing file:' print('Finished Writing file:'
, '\nFilename:', out_filename , '\nFilename:', outfile
, '\nPath:', outdir , '\n=============================================================')
, '\n=============================================================')
if __name__ == '__main__': if __name__ == '__main__':
main() main()
#%% end of script #%% end of script
#======================================================================= #=======================================================================