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
121 1.upto(m.size - 1).each do
122 nt = sam[:SEQ].seq[offset]
124 align << [offset, "#{nt}>-"]
131 nt2 = sam[:SEQ].seq[offset - deletions]
133 align << [offset, "#{nt}>#{nt2}"]
140 align.sort_by { |a| a.first }.map { |k, v| "#{k}:#{v}" }.join(",")
143 # Method to initialize a Sam object.
144 def initialize(io = nil)
152 @io.each_line do |line|
153 unless line[0] == '@'
154 entry = parse_alignment(line.chomp)
156 yield entry if block_given?
163 # Method to parse the header section of a SAM file.
164 # Each header line should match:
165 # /^@[A-Za-z][A-Za-z](\t[A-Za-z][A-Za-z0-9]:[ -~]+)+$/ or /^@CO\t.*/.
166 # Tags containing lowercase letters are reserved for end users.
168 @io.each_line do |line|
169 if line =~ /^@([A-Za-z][A-Za-z])/
175 when 'HD' then subparse_header(line)
176 when 'SQ' then subparse_sequence(line)
177 when 'RG' then subparse_read_group(line)
178 when 'PG' then subparse_program(line)
179 when 'CO' then subparse_comment(line)
181 raise SamError, "Unknown header tag: #{tag}"
189 return @header.empty? ? nil : @header
192 # Method to subparse header lines.
193 def subparse_header(line)
195 fields = line.split("\t")
197 if fields[1] =~ /^VN:([0-9]+\.[0-9]+)$/
200 raise SamError, "Bad version number: #{fields[1]}"
204 if fields[2] =~ /^SO:(unknown|unsorted|queryname|coordinate)$/
207 raise SamError, "Bad sort order: #{fields[2]}"
214 # Method to subparse sequence lines.
215 def subparse_sequence(line)
216 @header[:SQ] = Hash.new unless @header[:SQ].is_a? Hash
219 fields = line.split("\t")
221 if fields[1] =~ /^SN:([!-)+-<>-~][!-~]*)$/
224 raise SamError, "Bad sequence name: #{fields[1]}"
227 if fields[2] =~ /^LN:(\d+)$/
230 raise SamError, "Bad sequence length: #{fields[2]}"
233 (3 ... fields.size).each do |i|
234 if fields[i] =~ /^(AS|M5|SP|UR):([ -~]+)$/
237 raise SamError, "Bad sequence tag: #{fields[i]}"
241 @header[:SQ][:SN] = Hash.new unless @header[:SQ][:SN].is_a? Hash
243 if @header[:SQ][:SN].has_key? seq_name
244 raise SamError, "Non-unique sequence name: #{seq_name}"
246 @header[:SQ][:SN][seq_name] = hash
250 # Method to subparse read group lines.
251 def subparse_read_group(line)
252 @header[:RG] = Hash.new unless @header[:RG].is_a? Hash
255 fields = line.split("\t")
257 if fields[1] =~ /^ID:([ -~]+)$/
260 raise SamError, "Bad read group identifier: #{fields[1]}"
263 (2 ... fields.size).each do |i|
264 if fields[i] =~ /^(CN|DS|DT|FO|KS|LB|PG|PI|PL|PU|SM):([ -~]+)$/
267 raise SamError, "Bad read group tag: #{fields[i]}"
272 unless hash[:FO] =~ /^\*|[ACMGRSVTWYHKDBN]+$/
273 raise SamError, "Bad flow order: #{hash[:FO]}"
278 unless hash[:PL] =~ /^(CAPILLARY|LS454|ILLUMINA|SOLID|HELICOS|IONTORRENT|PACBIO)$/
279 raise SamError, "Bad platform: #{hash[:PL]}"
283 @header[:RG][:ID] = Hash.new unless @header[:RG][:ID].is_a? Hash
285 if @header[:RG][:ID].has_key? id
286 raise SamError, "Non-unique read group identifier: #{id}"
288 @header[:RG][:ID][id] = hash
292 # Method to subparse program lines.
293 def subparse_program(line)
294 @header[:PG] = Hash.new unless @header[:PG].is_a? Hash
297 fields = line.split("\t")
299 if fields[1] =~ /^ID:([ -~]+)$/
302 raise SamError, "Bad program record identifier: #{fields[1]}"
305 (2 ... fields.size).each do |i|
306 if fields[i] =~ /^(PN|CL|PP|VN):([ -~]+)$/
309 raise SamError, "Bad program record tag: #{fields[i]}"
313 @header[:PG][:ID] = Hash.new unless @header[:PG][:ID].is_a? Hash
315 if @header[:PG][:ID].has_key? id
316 raise SamError, "Non-unique program record identifier: #{id}"
318 @header[:PG][:ID][id] = hash
322 # Method to subparse comment lines.
323 def subparse_comment(line)
324 @header[:CO] = Array.new unless @header[:CO].is_a? Array
326 if line =~ /^@CO\t(.+)/
329 raise SamError, "Bad comment line: #{line}"
333 # Method to subparse alignment lines.
334 def parse_alignment(line)
335 fields = line.split("\t")
337 raise SamError, "Bad number of fields: #{fields.size} < 11" if fields.size < 11
340 flag = fields[1].to_i
343 mapq = fields[4].to_i
346 pnext = fields[7].to_i
347 tlen = fields[8].to_i
363 entry[:QNAME] = qname
364 entry[:FLAG] = Flag.new(flag)
365 entry[:RNAME] = rname
368 entry[:CIGAR] = Cigar.new(cigar)
369 entry[:RNEXT] = rnext
370 entry[:PNEXT] = pnext
372 entry[:SEQ] = (qual == '*') ? Seq.new(qname, seq) : Seq.new(qname, seq, qual)
375 # Optional fields - where some are really important! HATE HATE HATE SAM!!!
377 fields[11 .. -1].each do |field|
378 tag, type, val = field.split(':')
380 raise SamError, "Non-unique optional tag: #{tag}" if entry.has_key? tag.to_sym
382 # A [!-~] Printable character
384 # i [-+]?[0-9]+ Singed 32-bit integer
386 raise SamError, "Bad tag in optional field: #{field}" unless val =~ /^[-+]?[0-9]+$/
390 # f [-+]?[0-9]*\.?[0-9]+([eE][-+]?[0-9]+)? Single-precision floating number
391 # Z [ !-~]+ Printable string, including space
392 # H [0-9A-F]+ Byte array in the Hex format
393 # B [cCsSiIf](,[-+]?[0-9]*\.?[0-9]+([eE][-+]?[0-9]+)?)+ Integer or numeric array
395 entry[tag.to_sym] = val
401 # Method to check qname.
402 def check_qname(qname)
403 raise SamError, "Bad qname: #{qname}" unless qname =~ /^[!-?A-~]{1,255}$/
406 # Method to check flag.
408 raise SamError, "Bad flag: #{flag}" unless (0 .. 2**16 - 1).include? flag
411 # Method to check if rname, when not '*' and
412 # @SQ header lines are present, is located in
414 def check_rname(rname)
415 raise SamError, "Bad rname: #{rname}" unless rname =~ /^(\*|[!-()+-<>-~][!-~]*)$/
417 unless @header.empty? or rname == '*'
418 unless @header[:SQ][:SN].has_key? rname.to_sym
419 raise SamError, "rname not found in header hash: #{rname}"
424 # Method to check pos.
426 raise SamError, "Bad pos: #{pos}" unless (0 .. 2**29 - 1).include? pos
429 # Method to check mapq.
431 raise SamError, "Bad mapq: #{mapq}" unless (0 .. 2**8 - 1).include? mapq
434 # Method to check if rnext, when not '*' or '='
435 # and @SQ header lines are present, is located
436 # in the header hash.
437 def check_rnext(rnext)
438 raise SamError, "Bad rnext: #{rnext}" unless rnext =~ /^(\*|=|[!-()+-<>-~][!-~]*)$/
440 unless @header.empty? or rnext == '*' or rnext == '='
441 unless @header[:SQ][:SN].has_key? rnext.to_sym
442 raise SamError, "rnext not found in header hash: #{rnext}"
447 # Method to check pnext.
448 def check_pnext(pnext)
449 raise SamError, "Bad pnext: #{pnext}" unless (0 .. 2**29 - 1).include? pnext
452 # Method to check tlen.
454 raise SamError, "Bad tlen: #{tlen}" unless (-2**29 + 1 .. 2**29 - 1).include? tlen
457 # Method to check seq.
459 raise SamError, "Bad seq: #{seq}" unless seq =~ /^(\*|[A-Za-z=.]+)$/
462 # Method to check qual.
464 raise SamError, "Bad qual: #{qual}" unless qual =~ /^[!-~]+$/
467 # Method to deconvolute the SAM flag field.
471 # Method to initialize a Flag object.
476 # Method to test if template have
477 # multiple fragments in sequencing.
482 # Method to test if each fragment
483 # properly aligned according to the aligner.
488 # Method to test if the fragment was unmapped.
493 # Method to test if the next fragment was unmapped.
495 flag & FLAG_NEXT_UNMAPPED
498 # Method to test if the fragment was reverse complemented.
503 # Method to test if the next fragment was reverse complemented.
505 flag & FLAG_NEXT_REVCOMP
508 # Method to test if the fragment was first in the template.
513 # Method to test if the fragment was last in the template.
518 # Method to test for secondary alignment.
519 def secondary_alignment?
520 flag & FLAG_SECONDARY_ALIGNMENT
523 # Method to test for quality fail.
525 flag & FLAG_QUALITY_FAIL
528 # Method to test for PCR or optical duplicates.
530 flag & FLAG_DUPLICATES
536 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<