1 # Copyright (C) 2007-2011 Martin A. Hansen.
3 # This program is free software; you can redistribute it and/or
4 # modify it under the terms of the GNU General Public License
5 # as published by the Free Software Foundation; either version 2
6 # of the License, or (at your option) any later version.
8 # This program is distributed in the hope that it will be useful,
9 # but WITHOUT ANY WARRANTY; without even the implied warranty of
10 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11 # GNU General Public License for more details.
13 # You should have received a copy of the GNU General Public License
14 # along with this program; if not, write to the Free Software
15 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
17 # http://www.gnu.org/copyleft/gpl.html
19 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
21 # This software is part of the Biopieces framework (www.biopieces.org).
23 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
25 # SAM format version v1.4-r962 - April 17, 2011
27 # http://samtools.sourceforge.net/SAM1.pdf
30 require 'maasha/filesys'
32 require 'maasha/cigar'
34 REGEX_HEADER = Regexp.new(/^@[A-Za-z][A-Za-z](\t[A-Za-z][A-Za-z0-9]:[ -~]+)+$/)
35 REGEX_COMMENT = Regexp.new(/^@CO\t.*/)
37 FLAG_MULTI = 0x1 # Template having multiple fragments in sequencing
38 FLAG_ALIGNED = 0x2 # Each fragment properly aligned according to the aligner
39 FLAG_UNMAPPED = 0x4 # Fragment unmapped
40 FLAG_NEXT_UNMAPPED = 0x8 # Next fragment in the template unmapped
41 FLAG_REVCOMP = 0x10 # SEQ being reverse complemented
42 FLAG_NEXT_REVCOMP = 0x20 # SEQ of the next fragment in the template being reversed
43 FLAG_FIRST = 0x40 # The first fragment in the template
44 FLAG_LAST = 0x80 # The last fragment in the template
45 FLAG_SECONDARY_ALIGNMENT = 0x100 # Secondary alignment
46 FLAG_QUALITY_FAIL = 0x200 # Not passing quality controls
47 FLAG_DUPLICATES = 0x400 # PCR or optical duplicate
49 # Error class for all exceptions to do with Genbank.
50 class SamError < StandardError; end
52 # Class to parse and write SAM files.
54 attr_accessor :io, :header
56 # Class method to convert a SAM entry
57 # to a Biopiece record.
62 bp[:Q_ID] = sam[:QNAME]
63 bp[:STRAND] = sam[:FLAG].revcomp? ? '-' : '+'
64 bp[:S_ID] = sam[:RNAME]
65 bp[:S_BEG] = sam[:POS]
66 bp[:MAPQ] = sam[:MAPQ]
67 bp[:CIGAR] = sam[:CIGAR].to_s
69 unless sam[:RNEXT] == '*'
70 bp[:Q_ID2] = sam[:RNEXT]
71 bp[:S_BEG2] = sam[:PNEXT]
72 bp[:TLEN] = sam[:TLEN]
75 bp[:SEQ] = sam[:SEQ].seq
77 unless sam[:SEQ].qual.nil?
78 bp[:SCORES] = sam[:SEQ].convert_phred2illumina!.qual
81 if sam.has_key? :NM and sam[:NM].to_i > 0
82 bp[:ALIGN] = self.align_descriptors(sam)
88 # Class method to convert a Biopiece record
94 # Create alignment descriptors according to the KISS
96 # http://code.google.com/p/biopieces/wiki/KissFormat
97 def self.align_descriptors(sam)
102 sam[:CIGAR].each do |len, op|
104 (0 ... len).each_with_index do |i|
105 nt = sam[:SEQ].seq[offset + i]
107 align << [offset + i, "->#{nt}"]
117 sam[:MD].scan(/\d+|\^[A-Z]+|[A-Z]+/).each do |m|
118 if m =~ /\d+/ # Matches
120 elsif m[0] == '^' # Deletions
123 align << [offset, "#{nt}>-"]
130 nt2 = sam[:SEQ].seq[offset - deletions]
132 align << [offset, "#{nt}>#{nt2}"]
139 align.sort_by { |a| a.first }.map { |k, v| "#{k}:#{v}" }.join(",")
142 # Method to initialize a Sam object.
143 def initialize(io = nil)
151 @io.each_line do |line|
152 unless line[0] == '@'
153 entry = parse_alignment(line.chomp)
155 yield entry if block_given?
162 # Method to parse the header section of a SAM file.
163 # Each header line should match:
164 # /^@[A-Za-z][A-Za-z](\t[A-Za-z][A-Za-z0-9]:[ -~]+)+$/ or /^@CO\t.*/.
165 # Tags containing lowercase letters are reserved for end users.
167 @io.each_line do |line|
168 if line =~ /^@([A-Za-z][A-Za-z])/
174 when 'HD' then subparse_header(line)
175 when 'SQ' then subparse_sequence(line)
176 when 'RG' then subparse_read_group(line)
177 when 'PG' then subparse_program(line)
178 when 'CO' then subparse_comment(line)
180 raise SamError, "Unknown header tag: #{tag}"
188 return @header.empty? ? nil : @header
191 # Method to subparse header lines.
192 def subparse_header(line)
194 fields = line.split("\t")
196 if fields[1] =~ /^VN:([0-9]+\.[0-9]+)$/
199 raise SamError, "Bad version number: #{fields[1]}"
203 if fields[2] =~ /^SO:(unknown|unsorted|queryname|coordinate)$/
206 raise SamError, "Bad sort order: #{fields[2]}"
213 # Method to subparse sequence lines.
214 def subparse_sequence(line)
215 @header[:SQ] = Hash.new unless @header[:SQ].is_a? Hash
218 fields = line.split("\t")
220 if fields[1] =~ /^SN:([!-)+-<>-~][!-~]*)$/
223 raise SamError, "Bad sequence name: #{fields[1]}"
226 if fields[2] =~ /^LN:(\d+)$/
229 raise SamError, "Bad sequence length: #{fields[2]}"
232 (3 ... fields.size).each do |i|
233 if fields[i] =~ /^(AS|M5|SP|UR):([ -~]+)$/
236 raise SamError, "Bad sequence tag: #{fields[i]}"
240 @header[:SQ][:SN] = Hash.new unless @header[:SQ][:SN].is_a? Hash
242 if @header[:SQ][:SN].has_key? seq_name
243 raise SamError, "Non-unique sequence name: #{seq_name}"
245 @header[:SQ][:SN][seq_name] = hash
249 # Method to subparse read group lines.
250 def subparse_read_group(line)
251 @header[:RG] = Hash.new unless @header[:RG].is_a? Hash
254 fields = line.split("\t")
256 if fields[1] =~ /^ID:([ -~]+)$/
259 raise SamError, "Bad read group identifier: #{fields[1]}"
262 (2 ... fields.size).each do |i|
263 if fields[i] =~ /^(CN|DS|DT|FO|KS|LB|PG|PI|PL|PU|SM):([ -~]+)$/
266 raise SamError, "Bad read group tag: #{fields[i]}"
271 unless hash[:FO] =~ /^\*|[ACMGRSVTWYHKDBN]+$/
272 raise SamError, "Bad flow order: #{hash[:FO]}"
277 unless hash[:PL] =~ /^(CAPILLARY|LS454|ILLUMINA|SOLID|HELICOS|IONTORRENT|PACBIO)$/
278 raise SamError, "Bad platform: #{hash[:PL]}"
282 @header[:RG][:ID] = Hash.new unless @header[:RG][:ID].is_a? Hash
284 if @header[:RG][:ID].has_key? id
285 raise SamError, "Non-unique read group identifier: #{id}"
287 @header[:RG][:ID][id] = hash
291 # Method to subparse program lines.
292 def subparse_program(line)
293 @header[:PG] = Hash.new unless @header[:PG].is_a? Hash
296 fields = line.split("\t")
298 if fields[1] =~ /^ID:([ -~]+)$/
301 raise SamError, "Bad program record identifier: #{fields[1]}"
304 (2 ... fields.size).each do |i|
305 if fields[i] =~ /^(PN|CL|PP|VN):([ -~]+)$/
308 raise SamError, "Bad program record tag: #{fields[i]}"
312 @header[:PG][:ID] = Hash.new unless @header[:PG][:ID].is_a? Hash
314 if @header[:PG][:ID].has_key? id
315 raise SamError, "Non-unique program record identifier: #{id}"
317 @header[:PG][:ID][id] = hash
321 # Method to subparse comment lines.
322 def subparse_comment(line)
323 @header[:CO] = Array.new unless @header[:CO].is_a? Array
325 if line =~ /^@CO\t(.+)/
328 raise SamError, "Bad comment line: #{line}"
332 # Method to subparse alignment lines.
333 def parse_alignment(line)
334 fields = line.split("\t")
336 raise SamError, "Bad number of fields: #{fields.size} < 11" if fields.size < 11
339 flag = fields[1].to_i
342 mapq = fields[4].to_i
345 pnext = fields[7].to_i
346 tlen = fields[8].to_i
362 entry[:QNAME] = qname
363 entry[:FLAG] = Flag.new(flag)
364 entry[:RNAME] = rname
367 entry[:CIGAR] = Cigar.new(cigar)
368 entry[:RNEXT] = rnext
369 entry[:PNEXT] = pnext
371 entry[:SEQ] = (qual == '*') ? Seq.new(qname, seq) : Seq.new(qname, seq, qual)
374 # Optional fields - where some are really important! HATE HATE HATE SAM!!!
376 fields[11 .. -1].each do |field|
377 tag, type, val = field.split(':')
379 raise SamError, "Non-unique optional tag: #{tag}" if entry.has_key? tag.to_sym
381 # A [!-~] Printable character
383 # i [-+]?[0-9]+ Singed 32-bit integer
385 raise SamError, "Bad tag in optional field: #{field}" unless val =~ /^[-+]?[0-9]+$/
389 # f [-+]?[0-9]*\.?[0-9]+([eE][-+]?[0-9]+)? Single-precision floating number
390 # Z [ !-~]+ Printable string, including space
391 # H [0-9A-F]+ Byte array in the Hex format
392 # B [cCsSiIf](,[-+]?[0-9]*\.?[0-9]+([eE][-+]?[0-9]+)?)+ Integer or numeric array
394 entry[tag.to_sym] = val
400 # Method to check qname.
401 def check_qname(qname)
402 raise SamError, "Bad qname: #{qname}" unless qname =~ /^[!-?A-~]{1,255}$/
405 # Method to check flag.
407 raise SamError, "Bad flag: #{flag}" unless (0 .. 2**16 - 1).include? flag
410 # Method to check if rname, when not '*' and
411 # @SQ header lines are present, is located in
413 def check_rname(rname)
414 raise SamError, "Bad rname: #{rname}" unless rname =~ /^(\*|[!-()+-<>-~][!-~]*)$/
416 unless @header.empty? or rname == '*'
417 unless @header[:SQ][:SN].has_key? rname.to_sym
418 raise SamError, "rname not found in header hash: #{rname}"
423 # Method to check pos.
425 raise SamError, "Bad pos: #{pos}" unless (0 .. 2**29 - 1).include? pos
428 # Method to check mapq.
430 raise SamError, "Bad mapq: #{mapq}" unless (0 .. 2**8 - 1).include? mapq
433 # Method to check if rnext, when not '*' or '='
434 # and @SQ header lines are present, is located
435 # in the header hash.
436 def check_rnext(rnext)
437 raise SamError, "Bad rnext: #{rnext}" unless rnext =~ /^(\*|=|[!-()+-<>-~][!-~]*)$/
439 unless @header.empty? or rnext == '*' or rnext == '='
440 unless @header[:SQ][:SN].has_key? rnext.to_sym
441 raise SamError, "rnext not found in header hash: #{rnext}"
446 # Method to check pnext.
447 def check_pnext(pnext)
448 raise SamError, "Bad pnext: #{pnext}" unless (0 .. 2**29 - 1).include? pnext
451 # Method to check tlen.
453 raise SamError, "Bad tlen: #{tlen}" unless (-2**29 + 1 .. 2**29 - 1).include? tlen
456 # Method to check seq.
458 raise SamError, "Bad seq: #{seq}" unless seq =~ /^(\*|[A-Za-z=.]+)$/
461 # Method to check qual.
463 raise SamError, "Bad qual: #{qual}" unless qual =~ /^[!-~]+$/
466 # Method to deconvolute the SAM flag field.
470 # Method to initialize a Flag object.
475 # Method to test if template have
476 # multiple fragments in sequencing.
481 # Method to test if each fragment
482 # properly aligned according to the aligner.
487 # Method to test if the fragment was unmapped.
492 # Method to test if the next fragment was unmapped.
494 flag & FLAG_NEXT_UNMAPPED
497 # Method to test if the fragment was reverse complemented.
502 # Method to test if the next fragment was reverse complemented.
504 flag & FLAG_NEXT_REVCOMP
507 # Method to test if the fragment was first in the template.
512 # Method to test if the fragment was last in the template.
517 # Method to test for secondary alignment.
518 def secondary_alignment?
519 flag & FLAG_SECONDARY_ALIGNMENT
522 # Method to test for quality fail.
524 flag & FLAG_QUALITY_FAIL
527 # Method to test for PCR or optical duplicates.
529 flag & FLAG_DUPLICATES
535 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<