creating tab-delimited output from unix

[tonybert]

greetings, i have a large list of taxonomic information from a blast search in a text file (taxa.txt, see below).

$ head -5 taxa.txt
Eukaryota;Opisthokonta;Fungi;Ascomycota
Bacteria;Candidate division OD1;marine metagenome
Bacteria;Planctomycetes;Phycisphaerae
Bacteria;Cyanobacteria;Chloroplast;
Eukaryota;Opisthokonta;Fungi;Ascomycota;

What i would like to do is create a bar plot of this data in excel or R. To do this, I have been using sort and uniq to tally up each of the hits.

$sort -u taxa.txt|uniq -c > sorted_taxa.txt
$head -5 sorted_taxa.txt
1 Archaea;Euryarchaeota;Halobacteria
1 Archaea;Euryarchaeota;Thermoplasmata
1 Archaea;Thaumarchaeota;Marine Group I
1 Archaea;Thaumarchaeota;Marine Group I
1 Archaea;Thaumarchaeota;Soil Crenarchaeotic Group(SCG)

However, when i open this file, and copy and paste the results into excel, it treats each row as one single string. What I would like is for there to be two columns (at least) in output, the column describing the count, and the column describing the taxonomic information. I believe this would be tab-delimited output. Is there a way to tell unix to produce such a result? Thanks,

-Tony

[BAMseek]

Hi Tony,

First thing I noticed is that I think you want to remove the "-u" flag from the sort. The -u flag will remove duplicate lines, so when you do "uniq -c", you will get all 1s.

Something like this might work:

sort taxa.txt | uniq -c | awk '{$1=$1};1' | awk '{sub(/\ /,"\t")};1' > histo.txt

The first awk command gets rid of leading, trailing, and extra white spaces. The second awk command replaces the first space with a tab (both commands from awk one-liners).

Hope that helps,
Justin

[maubp]

Why not open it in Excel (or R or ...) but specify the separators are semi-colons?