Re: New to programming need help for inserting text

[kapoormanav]

Hi all

I am new to programming. I have two very big files

File 1: (>80,000 lines)

CCDS635.1_0 123 G S 255 33 1.00 63 60 G 255 C
CCDS635.1_0 175 C S 255 51 0.94 63 62 C 243 G
CCDS635.1_8 259 G R 90 254 1.00 63 62 G 255 A
CCDS635.1_14 328 A T 39 4 0.00 63 36 N 158 N
CCDS635.1_16 80 G K 139 22 1.00 63 62 G 255 T

File 2:

CCDS635.1_0 1 67162857
CCDS635.1_1 1 67143396
CCDS635.1_2 1 67131424
CCDS635.1_3 1 67129349
CCDS635.1_4 1 67112976

I want to insert all three columns of File 2 to File one at the
respective positions according to column 1 of File 1.

e.g.

File 3 should be

CCDS635.1_0 1 67162857 123 G S 255 33 1.00 63 60 G 255 C
CCDS635.1_0 1 67162857 175 C S 255 51 0.94 63 62 C 243 G

Can anyone help me with this?

Manav

[adamdeluca]

join 2.txt 1.txt > 3.txt

on mac/linux

[kapoormanav]

Thanks for replying.
I tried to join but it gives error:

join: file 1 is not in sorted order
join: file 2 is not in sorted order

There is another problem also..These both files have different number of lines.
I don't want exactly merging of file, I want to create new file where
the similar identifier get replaced by three columns of 2nd file. If
similar identifier occurs in 4 rows of 1st file then the next 2
columns of the same identifier in file 2 get inserted before 2nd
column of file one for all 4 rows. It can be like find and replace
where identifier "CCDS635.1_0" will be replaced by "CCDS635.1_0
1 67162857" in all rows. I can do it with "sed" command of unix
but then "sed" will do one identifier at one time and I have >1000.
And everytime sed will create new file. So I was looking for solution
where "identifier" replaced by new identifier with 3 columns.

Manav

[kapoormanav]

Thanks...It worked I given the option join --nocheck-order

Manav

[Jose Blanca]

Maybe somthing like this would work:

#!/bin/env python
   
def merge_files(fname1, fname2, out_fname):
  
     #read file 2 into memory
     file2 = {}
     for line in open(fname2):
         line = line.strip()
         if not line:
             continue
         line_name = line.split()[0]
         file2[line_name] = line
 
     #now create the output
     output_fhand = open(out_fname, 'w')
     for line in open(fname1):
         line = line.strip()
         line_name, line = line.split(' ', 1)
         line_name = file2[line_name]
         output_fhand.write(' '.join((line_name, line, '\n')))
   
   if __name__ == '__main__':
       merge_files('file1.txt', 'file2.txt', 'merged.txt')

[kapoormanav]

Thanks for the help.

awk 'BEGIN{FS=OFS="\t"}FNR==NR{a[$1]=$2}FNR!=NR{print $1,a[$1],a[$2],$2,$3,$4,$5,$6,$7,$8,$9}' exonlegth.txt coor.txt > coor1.txt

worked for me and I was able to join the two files without sorting.

Thanks all for help

Manav