Sha256: 4471d7b1c211b6d9b051efc744ed7a424e35427c32f936d5e1de222e644461d7
Contents?: true
Size: 1.42 KB
Versions: 13
Compression:
Stored size: 1.42 KB
Contents
##
# SQUISH A CSV FILE BY COLUMN NAME
#
module Lederhosen
class CLI
desc 'squish', 'merge cell values (reads) in a csv file by column name (cluster)'
method_option :csv_file, :type => :string, :required => true
method_option :output, :type => :string, :required => false
def squish
csv_file = options[:csv_file]
output = options[:output] || $stdout
# sample_name -> column name -> total number of reads
total_by_sample_by_column = Hash.new { |h, k| h[k] = Hash.new { |h, k| h[k] = 0 } }
column_names = '' # scope
# Load CSV file, merge counts in columns with the same name
File.open(csv_file) do |handle|
column_names = handle.gets.strip.split(',')[1..-1]
handle.each do |line|
line = line.strip.split(',')
sample = line[0]
line[1..-1].zip(column_names) do |reads, column_name|
total_by_sample_by_column[sample][column_name] += reads.to_i
end
end
end
output = File.open(output, 'w') rescue $stdout
# print the new, squished csv file
column_names.uniq!.sort!
output.puts "-,#{column_names.join(',')}"
total_by_sample_by_column.each_pair do |sample_id, row|
output.print "#{sample_id}"
column_names.each do |column_name|
output.print ",#{row[column_name]}"
end
output.print "\n"
end
output.close
end
end
end
Version data entries
13 entries across 13 versions & 1 rubygems