I have more than 100 tab delimited files with seven different columns in each files. I was to get a single file with first and 7th column of first file and seventh columns of every other files. Any help will be appreciated.
file1:
A B C D E F G
gene1 6 12 3 4 0 1
gene2 8 0 2 5 10 3
gene3 9 1 1 6 11 5
gene4 10 3 0 7 2 7
file 2:
A H I J K K M
gene1 0 2 1 2 4 2
gene2 10 3 5 5 7 4
gene3 20 40 7 0 9 6
gene4 1 25 9 2 10 7
file3:
file4: so on
I want to get output like this:
A G M
gene1 1 2
gene2 3 4
gene3 5 6
gene4 7 7
2 answers
paste <(cut -f1,7 file1) <(cut -f7 file2)
Assuming all files end on .txt using GNU parallel:
paste <(cut -f1,7 file1.txt) <(tr "\n" "\t" < <(ls *.txt | grep -v 'file1.txt' | sort -V | parallel -k "cut -f7 {}"))
or with a for loop:
paste <(cut -f1,7 file1.txt) <(tr "\n" "\t" < <(for i in `ls *.txt | sort -V`; do if [[ $i == "file1.txt" ]]; then continue; fi; cut -f7 $i; done))
Perhaps worth noting that you can paste as many columns as you want (up to the limit on command line length), e.g.:
$ paste <(cut -f1,7 file1) <(cut -f7 file2) <(...) <(cut -f7 fileN) > answer.txt
To do this programmatically, you might use a script to generate the <(cut -f7 fileX) statements for files 2 through N.
with datamash (present in most of the distro repos): (note: Do not output to a .txt file)
output;
$ paste *.txt | datamash transpose | awk 'NR==1{print};NR%7==0{print}' | datamash transpose
A G M
gene1 1 2
gene2 3 4
gene3 5 6
gene4 7 7
or
$ cut -f1,7 *.txt | datamash -sg 1 collapse 2 | tr -s "," "\t"
A G M
gene1 1 2
gene2 3 4
gene3 5 6
gene4 7 7
input:
$ tail -n+2 *.txt
==> file1.txt <==
gene1 6 12 3 4 0 1
gene2 8 0 2 5 10 3
gene3 9 1 1 6 11 5
gene4 10 3 0 7 2 7
==> file2.txt <==
gene1 0 2 1 2 4 2
gene2 10 3 5 5 7 4
gene3 20 40 7 0 9 6
gene4 1 25 9 2 10 7
Log in to answer this question.
-
Understanding Uncommon Genotype Formats in VCF Files Generated by nf-core RNAvar Pipeline
written by ASid •
-
filtering out lines (rows) based on repeated zero values with awk in linux
written by rgescudero •
-
Specification regarding paralogs
written by K.Gee •
-
How to multiple apply continuous mapping and Discrete mapping each column in Cytoscape
written by mmh7272
-
Count columns and delet specific one
written by salvatore_d •
-
genie3 package Error in (function (classes, fdef, mtable)
written by Sam
-
RIblast output Circos plot
written by ta_awwad
-
Count non-zero elements along the column and append to table in R
written by User000
-
how to perform pairwise correlation coefficent in r program for gene expression matrix
written by rajasekargutha •
-
Filteration of 0 counts in all column of the htseq-count metrics and add pseudo counts in remaining…
written by bioinforupesh2009.au •
Please edit your question and explain how this is related to bioinformatics. If not, the post will be closed as off-topic.
If the first column of the input files are not same, paste command will give wrong results
This comment is not an answer by itself, but a comment on an existing answer. I've moved it to a comment on the top level post.
The question clearly shows that all files share the same first column. If that were not to be the case, OP would state that.