Created
August 16, 2012 04:33
-
-
Save shujishigenobu/3366873 to your computer and use it in GitHub Desktop.
Reading tab-delimited tables, retrieve taxonomy information. It works only with NCBI nr database.
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
| #=== | |
| # nr_taxinfo.rb | |
| # | |
| # Reading tab-delimited tables, retrieve taxonomy information. | |
| # It works only with NCBI nr database. | |
| # Three columns, tax_id, scientific_name and common_name, are added. | |
| $infile = ARGV[0] | |
| $blastdb = ARGV[1] | |
| $keypos = (ARGV[2] || 1).to_i | |
| cache = Hash.new | |
| File.open($infile).each do |l| | |
| next if /^#/.match(l) | |
| a = l.chomp.split(/\t/) | |
| target_id = a[$keypos] | |
| # p target_id | |
| unless cache[target_id] | |
| cmd = "fastacmd -d #{$blastdb} -s \"#{target_id}\" -T" | |
| # puts cmd | |
| res = nil | |
| IO.popen(cmd){|io| res = io.read} | |
| # puts res | |
| records = res.split(%r{\n\n}) | |
| r = records.first | |
| ## analyze only first record | |
| ## TODO: should be analyzed multiple records | |
| if /^NCBI/.match(r) | |
| tid = /^NCBI taxonomy id:\s*(\d+)/.match(r)[1] | |
| tid = tid.to_i | |
| cname = /^Common name:\s*(.+)\n/.match(r)[1] | |
| sname = /^Scientific name:\s*(.+)$/.match(r)[1] | |
| cache[target_id] = {'tid' => tid, 'cname' => cname, 'sname' => sname} | |
| else | |
| ## malformed report | |
| STDERR.puts "WARNING: taxinfo not found for #{target_id}" | |
| cache[target_id] = {'tid' => nil, 'cname' => nil, 'sname' => nil} | |
| end | |
| end | |
| data = cache[target_id] | |
| outdat = a | |
| outdat += [data['tid'], data['sname'], data['cname']] | |
| puts outdat.flatten.join("\t") | |
| end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment