[SOLVED] Stuck !!!!

wackysiya · 03-12-2013, 01:06 AM

I have a sequence :

Code:

PH01000000G0240 P.he_genemodel_v1.0 CDS 120721 121773 . - . ID=PH01000000G0240.CDS;Parent=PH01000000G0240
PH01000001G0190 P.he_genemodel_v1.0 mRA 136867 137309 . - . ID=PH01000001G0190.mRNA;Parent=PH01000001G0190
.............................................
PH01278028G0010 P.he_genemodel_v1.0 CDS 27 501.. . - . ID=PH01278028G0010;Description="oereed"
PH01278104G0010 P.he_genemodel_v1.0 CDS 34 171 . - . ID=PH01278104G0010.CDS;Parent=PH01278104G0010

I want to replace PH0100000 by string but only in the first tab like
PH01000000 to string0
PH01000001 to string1
....
PH01278104 to string278104
PH01278028 to string278028

I want it to look like

Code:

string0G0240 P.he_genemodel_v1.0 CDS 120721 121773 . - . ID=PH01000000G0240.CDS;Parent=PH01000000G0240
string1G0190 P.he_genemodel_v1.0 mRA 136867 137309 . - . ID=PH01000001G0190.mRNA;Parent=PH01000001G0190
.............................................

string278028G0010 P.he_genemodel_v1.0 CDS 27 501.. . - . ID=PH01278028G0010;Description="oereed"
string278104G0010 P.he_genemodel_v1.0 CDS 34 171 . - . ID=PH01278104G0010.CDS;Parent=PH01278104G0010

I used sed command but it didnt work.
sed 's/^PH01\t/string\t/' infile > sortedfile
and then
sed 's/string0*/string/g' infile>sorted file

so tht it removes the extra 00 formed

so that any replacement of this form

PH01000000 to string00000

gets converted to this form
PH01000000 to string0

But none of my commands worked

Could some1 please help

shivaa · 03-12-2013, 02:10 AM

You can try awk, as:

Code:

~$ gawk -F" " '{gsub(/PH01278104/,"<string>",$1); print $0}' infile.txt

Code:

~$ nawk -F" " '{gsub(/PH01278104/,"<string>",$1); print $0}' infile.txt

wackysiya · 03-12-2013, 10:38 AM

I used $ gawk -F" " '{gsub(/PH01/,"string",$1); print $0}' infile.txt

It replaced all my entries by string
PH01000000G124 to string00000G124
PH01000000G144 to string00000G144
...
PH01000001G670 to string00001G670
PH01000002G450 to string00001G450

PH01278028G433 to string278028G433
THE LAST ENTRY AS IT IS ACCORDING TO MY REQUIREMENT OF STRING FOLLOWED BY NUMBER

BUT THE OTHER ENTRIES HAVE ADDITIONAL ZEROES.
PH01000000G124 to string00000G124
I WANT IT LIKE string0G124 ,string1G453

Is there anythng i could use that i could get ride of that extra zeroes in the first tab.

I used sed 's/string0*/string/g' infile>sorted file but it didnt work.

Thanks a lot!!!

grail · 03-12-2013, 11:14 AM

Try just sub as gsub has the 'g' for global on the line.

shivaa · 03-12-2013, 11:15 AM

Let's know exactly how many such entries are there in your infile.txt ?
For 2 or 3 entries, it's ok to use gsub function in awk, else we need to use some regexp to replace multiple entries.

wackysiya · 03-12-2013, 12:10 PM

Yes , i have lot of entries like 10000 .

using gsub i got the first part done.

I want to now replace

string00000G670 to string0G670
string00001G270 to string1G270
string00002G634 to string2G634

THIS MODIFICATION HAS TO BEEN ONLY IN THE FIRST TAB.

I used sed but it didnt work
Could you suggest?

shivaa · 03-12-2013, 12:23 PM

You can use the same here as:

Code:

~$ gawk -F" " '{sub(/string0000/,"string",$1); print $0}' infile.txt

OR, using sed:

Code:

~$ sed -e 's/string0000/string/' infile.txt

wackysiya · 03-12-2013, 03:44 PM

It worked thanks!!!