CSV to PLINK

: csv to (map, ped)

• CSV format : columns=[CHR, id, pheno, variant1, variant2,...]

  • CHR : chromosome
  • pheno : case 1 / control 0

#!/usr/bin/env python

import sys

infile_name = sys.argv[1]

pedDict = {
    "0" : "0 0",
    "1" : "A A",
    "2" : "A a",
    "3" : "a a"
}

def convertToPlink(infile_name):
    with open(infile_name, 'r') as infile:
        header = infile.readline().rstrip().split()
        chromosome = infile.readline().split()[0]
        with open('test' + chromosome + '.map', 'w') as mapfile:
            a=1
            for POS in header[3:]:
                mapfile.write("\t".join([chromosome,POS, "0", str(a)])+"\n")
                a+=1
    with open(infile_name, 'r') as infile:
        with open('test' + chromosome + '.ped', 'w') as pedfile:
            id_index = 0
            for line in infile:
                if not line.startswith("CHR"):
                    id_index += 1
                    ID = "i_" + str(id_index)
                    line = line.rstrip().split()
                    IID= line[1]
                    pheno = str(int(line[2])+1)

                    pedfile.write(" ".join([ID, IID, "0", "0", "0",pheno]+[pedDict[genotype] for genotype in line[3:]])+ "\n")

convertToPlink(infile_name)

Include charactors or words

df=pd.read_csv('mydata.csv')
col=df.columns

searchfor=['hel','good']
s1=col[col.str.contains('|'.join(searchfor))]

print(s1.values)

['hello', 'hellgate', 'good_morning', 'good_afternoon']

Split by separator

a=[['asd/123',34],['qwe/234',45],['zxc/456',67],['fgh/678',89]]

df=pd.DataFrame(a)

print(df)
print()
print(df[0].str.split('/'))
print(df[0].str.split('/').shape)
print(df[0].str.split('/',expand=True))
print(df[0].str.split('/',expand=True).shape)

         0   1
0  asd/123  34
1  qwe/234  45
2  zxc/456  67
3  fgh/678  89

0    [asd, 123]
1    [qwe, 234]
2    [zxc, 456]
3    [fgh, 678]
Name: 0, dtype: object
(4,)
     0    1
0  asd  123
1  qwe  234
2  zxc  456
3  fgh  678
(4, 2)

Merge two dataframes

df_final=pd.merge(df_left,df_right,left_index=True, right_index=True,how='left')

Contains value

import pandas as pd

df=pd.read_csv('mydata.csv',sep='\t+|\s+',header=None)

col=df[1]

s1=col[col.str.contains('.1', regex=False)]

print(len(s1))

String to number in DF

df=df.apply(pd.to_numeric, errors='ignore')

Merge dataframes

df2=pd.merge(df,df1, how='outer', left_index=True, right_index=True)

Display all

with pd.option_context('display.max_rows', None, 'display.max_columns', None):
    print(df)

Separate ID

import pandas as pd

df =pd.read_csv('list',sep='/',header=None)
df0=df[5].str.split('.',expand=True)[0]
step=100

for j in range(0,df.shape[0],step):
    if j== df.shape[0]//step*step:
        step=df.shape[0]%step
    ini=j
    fin=ini+step

    df1=pd.DataFrame(['hs38DH_re.txt'])
    df1=df1.append([step], ignore_index=True)

    for i in enumerate(df[5].tolist()):
        if (i[0]>=ini) and (i[0]<fin):
            df1=df1.append([i[1]], ignore_index=True)

    for i in enumerate(df0.tolist()):
        if (i[0]>=ini) and (i[0]<fin):
            df1=df1.append([i[1]], ignore_index=True)

    print(df1)

    #print(df1[0].to_string(index=False))
    df1.to_csv('In_File_'+str(ini)+'.txt',index=None,header=None)

Select data

import sys
import time

def help():
    print("Usage: python {} [ chrn ] \n".format(sys.argv[0]))
    exit()

# decorator
def runTime(func):
    def wrapper(*args, **kwargs):
        start = time.time()
        result = func(*args, **kwargs)
        end = time.time()
        print('# module : {0} / time(sec) : {1}'.format(func.__name__, end - start))
        return result
    return wrapper

@runTime
def readFiles():
    inFile=open(inf,'r')
    outFile=open(out,'w')

    bedFile=open(inbed,'r')
    a0='1 0'
    while 1:
        a=bedFile.readline()
        b='1 0'
        if not a :break
        if a.split()[1]==a0.split()[1]:continue
        a0=a
        while a0.split()[1]!=b.split()[1]:
            b=inFile.readline()

        outFile.write(b)

if __name__ == "__main__":
    if not(len(sys.argv) == 2):
        help()
    chrn = sys.argv[1]

    # 
    inf="Out_chr{}.txt".format(chrn)
    out="Out_chr{}.sel".format(chrn)

    # selection range
    inbed = "selected_range_chr{}.sel".format(chrn)  

    fadata = [0]
    bedData = []

    readFiles()

    sys.exit(0)

ADD

def plink_cmd(outputFile):
    inputFile=outputFile+'_ibd'

    df=pd.read_table(outputFile+'.ann',header=None)

    def add_gene(header,col,tempExt,outputExt):
        cmd='head -n 1 '+outputFile+'.'+tempExt
        lh = len(subprocess.check_output(cmd,shell=True,encoding='utf-8').split())

        cmd='echo '+header+' > '+outputFile+'.tmp && awk \'{print $'+str(col)+'}\' '+outputFile+'.'+tempExt
        snps=subprocess.check_output(cmd,shell=True,encoding='utf-8').split('\n')[1:-1]

        genes=parmap.map(find_gene,snps,df,pm_pbar=True,pm_processes=40)

        with open(outputFile+'.tmp','a') as f:
            for i in genes:
                f.write(i+'\n')

        cols='\"\\t\"'.join(map(lambda x:'$'+str(x),range(1,col+1)))
        endCols='\"\\t\"'.join(map(lambda x:'$'+str(x),range(col+1,lh+1)))

        if endCols=='':
            cmd='awk \'{print '+cols+'}\' '+outputFile+'.'+tempExt+'|paste - '+outputFile+'.tmp >'+outputFile+'.'+outputExt

        else:
            cmd='awk \'{print '+cols+'}\' '+outputFile+'.'+tempExt+'|paste - '+outputFile+'.tmp <(awk \'{print '+endCols+'}\' '+outputFile+'.'+tempExt+') >'+outputFile+'.'+outputExt

        subprocess.run(cmd,shell=True, executable="/bin/bash", stdout=subprocess.DEVNULL)

    add_gene('GENE',3,'txt','reg')