Created
August 16, 2012 04:58
-
-
Save shujishigenobu/3367058 to your computer and use it in GitHub Desktop.
Obtain full path of lineage from root to target species
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #=== | |
| # taxpath.rb | |
| # Obtain full path of lineage from root to target species | |
| ### conf ### | |
| taxtab = "/home/DB/public/processed/NCBI/taxonomy/taxdump/nodes.dmp" | |
| nametab = "/home/DB/public/processed/NCBI/taxonomy/taxdump/names.dmp" | |
| ### | |
| infile = ARGV[0] | |
| tid_position = (ARGV[1] || 0).to_i | |
| rel = Hash.new | |
| File.open(taxtab).each do |l| | |
| a = l.chomp.split(/\|/).map{|x| x.strip} | |
| rel[a[0].to_i] = a[1].to_i | |
| end | |
| name = {} | |
| File.open(nametab).each do |l| | |
| a = l.chomp.split(/\|/).map{|x| x.strip} | |
| next unless a[3] == "scientific name" | |
| name[a[0].to_i] = a[1] | |
| end | |
| File.open(infile).each do |l| | |
| path = [] | |
| a = l.chomp.split(/\t/) | |
| tid = a[tid_position] | |
| if /\d+/.match(tid) | |
| q = tid.to_i | |
| path << q | |
| loop do | |
| r = rel[q] | |
| path << r | |
| break if q == 1 | |
| q = r | |
| end | |
| path.pop | |
| path_txt = path.map{|x| name[x]} | |
| out_path = path.reverse.join(":") | |
| out_path_txt = path_txt.reverse.join(":") | |
| else | |
| out_path = nil | |
| out_path_txt = nil | |
| end | |
| out = [] | |
| out += a | |
| out << out_path | |
| out << out_path_txt | |
| puts out.join("\t") | |
| end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment