1 # Copyright (C) 2007-2013 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[:S_END] = sam[:POS] + sam[:SEQ].length - 1
67 bp[:MAPQ] = sam[:MAPQ]
68 bp[:CIGAR] = sam[:CIGAR].to_s
70 unless sam[:RNEXT] == '*'
71 bp[:Q_ID2] = sam[:RNEXT]
72 bp[:S_BEG2] = sam[:PNEXT]
73 bp[:TLEN] = sam[:TLEN]
76 bp[:SEQ] = sam[:SEQ].seq
78 unless sam[:SEQ].qual.nil?
79 bp[:SCORES] = sam[:SEQ].qual
82 if sam[:NM] and sam[:NM].to_i > 0
85 bp[:ALIGN] = self.align_descriptors(sam)
91 # Class method to create a new SAM entry from a Biopiece record.
100 rnext = bp[:Q_ID2] || '*'
101 pnext = bp[:S_BEG2] || 0
102 tlen = bp[:TLEN] || 0
104 qual = bp[:SCORES] || '*'
105 nm = "NM:i:#{bp[:NM]}" if bp[:NM]
106 md = "MD:Z:#{bp[:MD]}" if bp[:MD]
108 flag |= FLAG_REVCOMP if bp[:STRAND] == '+'
110 if qname && flag && rname && pos && mapq && cigar && rnext && pnext && tlen && seq && qual
111 ary = [qname, flag, rname, pos, mapq, cigar, rnext, pnext, tlen, seq, qual]
119 # Create alignment descriptors according to the KISS
120 # format description:
121 # http://code.google.com/p/biopieces/wiki/KissFormat
122 def self.align_descriptors(sam)
127 sam[:CIGAR].each do |len, op|
129 (0 ... len).each_with_index do |i|
130 nt = sam[:SEQ].seq[offset + i]
132 align << [offset + i, "->#{nt}"]
142 sam[:MD].scan(/\d+|\^[A-Z]+|[A-Z]+/).each do |m|
143 if m =~ /\d+/ # Matches
145 elsif m[0] == '^' # Deletions
148 align << [offset, "#{nt}>-"]
155 nt2 = sam[:SEQ].seq[offset - deletions]
157 align << [offset, "#{nt}>#{nt2}"]
164 align.sort_by { |a| a.first }.map { |k, v| "#{k}:#{v}" }.join(",")
167 # Method to initialize a Sam object.
168 def initialize(io = nil)
176 @io.each_line do |line|
177 unless line[0] == '@'
178 entry = parse_alignment(line.chomp)
180 yield entry if block_given?
187 # Method to parse the header section of a SAM file.
188 # Each header line should match:
189 # /^@[A-Za-z][A-Za-z](\t[A-Za-z][A-Za-z0-9]:[ -~]+)+$/ or /^@CO\t.*/.
190 # Tags containing lowercase letters are reserved for end users.
192 @io.each_line do |line|
193 if line =~ /^@([A-Za-z][A-Za-z])/
199 when 'HD' then subparse_header(line)
200 when 'SQ' then subparse_sequence(line)
201 when 'RG' then subparse_read_group(line)
202 when 'PG' then subparse_program(line)
203 when 'CO' then subparse_comment(line)
205 raise SamError, "Unknown header tag: #{tag}"
213 return @header.empty? ? nil : @header
216 # Method to subparse header lines.
217 def subparse_header(line)
219 fields = line.split("\t")
221 if fields[1] =~ /^VN:([0-9]+\.[0-9]+)$/
224 raise SamError, "Bad version number: #{fields[1]}"
228 if fields[2] =~ /^SO:(unknown|unsorted|queryname|coordinate)$/
231 raise SamError, "Bad sort order: #{fields[2]}"
238 # Method to subparse sequence lines.
239 def subparse_sequence(line)
240 @header[:SQ] = Hash.new unless @header[:SQ].is_a? Hash
243 fields = line.split("\t")
245 if fields[1] =~ /^SN:([!-)+-<>-~][!-~]*)$/
248 raise SamError, "Bad sequence name: #{fields[1]}"
251 if fields[2] =~ /^LN:(\d+)$/
254 raise SamError, "Bad sequence length: #{fields[2]}"
257 (3 ... fields.size).each do |i|
258 if fields[i] =~ /^(AS|M5|SP|UR):([ -~]+)$/
261 raise SamError, "Bad sequence tag: #{fields[i]}"
265 @header[:SQ][:SN] = Hash.new unless @header[:SQ][:SN].is_a? Hash
267 if @header[:SQ][:SN][seq_name]
268 raise SamError, "Non-unique sequence name: #{seq_name}"
270 @header[:SQ][:SN][seq_name] = hash
274 # Method to subparse read group lines.
275 def subparse_read_group(line)
276 @header[:RG] = Hash.new unless @header[:RG].is_a? Hash
279 fields = line.split("\t")
281 if fields[1] =~ /^ID:([ -~]+)$/
284 raise SamError, "Bad read group identifier: #{fields[1]}"
287 (2 ... fields.size).each do |i|
288 if fields[i] =~ /^(CN|DS|DT|FO|KS|LB|PG|PI|PL|PU|SM):([ -~]+)$/
291 raise SamError, "Bad read group tag: #{fields[i]}"
296 unless hash[:FO] =~ /^\*|[ACMGRSVTWYHKDBN]+$/
297 raise SamError, "Bad flow order: #{hash[:FO]}"
302 unless hash[:PL] =~ /^(CAPILLARY|LS454|ILLUMINA|SOLID|HELICOS|IONTORRENT|PACBIO)$/
303 raise SamError, "Bad platform: #{hash[:PL]}"
307 @header[:RG][:ID] = Hash.new unless @header[:RG][:ID].is_a? Hash
309 if @header[:RG][:ID][id]
310 raise SamError, "Non-unique read group identifier: #{id}"
312 @header[:RG][:ID][id] = hash
316 # Method to subparse program lines.
317 def subparse_program(line)
318 @header[:PG] = Hash.new unless @header[:PG].is_a? Hash
321 fields = line.split("\t")
323 if fields[1] =~ /^ID:([ -~]+)$/
326 raise SamError, "Bad program record identifier: #{fields[1]}"
329 (2 ... fields.size).each do |i|
330 if fields[i] =~ /^(PN|CL|PP|VN):([ -~]+)$/
333 raise SamError, "Bad program record tag: #{fields[i]}"
337 @header[:PG][:ID] = Hash.new unless @header[:PG][:ID].is_a? Hash
339 if @header[:PG][:ID][id]
340 raise SamError, "Non-unique program record identifier: #{id}"
342 @header[:PG][:ID][id] = hash
346 # Method to subparse comment lines.
347 def subparse_comment(line)
348 @header[:CO] = Array.new unless @header[:CO].is_a? Array
350 if line =~ /^@CO\t(.+)/
353 raise SamError, "Bad comment line: #{line}"
357 # Method to subparse alignment lines.
358 def parse_alignment(line)
359 fields = line.split("\t")
361 raise SamError, "Bad number of fields: #{fields.size} < 11" if fields.size < 11
364 flag = fields[1].to_i
367 mapq = fields[4].to_i
370 pnext = fields[7].to_i
371 tlen = fields[8].to_i
387 entry[:QNAME] = qname
388 entry[:FLAG] = Flag.new(flag)
389 entry[:RNAME] = rname
392 entry[:CIGAR] = Cigar.new(cigar)
393 entry[:RNEXT] = rnext
394 entry[:PNEXT] = pnext
396 entry[:SEQ] = (qual == '*') ? Seq.new(seq_name: qname, seq: seq) : Seq.new(seq_name: qname, seq: seq, qual: qual)
399 # Optional fields - where some are really important! HATE HATE HATE SAM!!!
401 fields[11 .. -1].each do |field|
402 tag, type, val = field.split(':')
404 raise SamError, "Non-unique optional tag: #{tag}" if entry[tag.to_sym]
406 # A [!-~] Printable character
408 # i [-+]?[0-9]+ Singed 32-bit integer
410 raise SamError, "Bad tag in optional field: #{field}" unless val =~ /^[-+]?[0-9]+$/
414 # f [-+]?[0-9]*\.?[0-9]+([eE][-+]?[0-9]+)? Single-precision floating number
415 # Z [ !-~]+ Printable string, including space
416 # H [0-9A-F]+ Byte array in the Hex format
417 # B [cCsSiIf](,[-+]?[0-9]*\.?[0-9]+([eE][-+]?[0-9]+)?)+ Integer or numeric array
419 entry[tag.to_sym] = val
425 # Method to check qname.
426 def check_qname(qname)
427 raise SamError, "Bad qname: #{qname}" unless qname =~ /^[!-?A-~]{1,255}$/
430 # Method to check flag.
432 raise SamError, "Bad flag: #{flag}" unless (0 .. 2**16 - 1).include? flag
435 # Method to check if rname, when not '*' and
436 # @SQ header lines are present, is located in
438 def check_rname(rname)
439 raise SamError, "Bad rname: #{rname}" unless rname =~ /^(\*|[!-()+-<>-~][!-~]*)$/
441 unless @header.empty? or rname == '*'
442 unless @header[:SQ][:SN][rname.to_sym]
443 raise SamError, "rname not found in header hash: #{rname}"
448 # Method to check pos.
450 raise SamError, "Bad pos: #{pos}" unless (0 .. 2**29 - 1).include? pos
453 # Method to check mapq.
455 raise SamError, "Bad mapq: #{mapq}" unless (0 .. 2**8 - 1).include? mapq
458 # Method to check if rnext, when not '*' or '='
459 # and @SQ header lines are present, is located
460 # in the header hash.
461 def check_rnext(rnext)
462 raise SamError, "Bad rnext: #{rnext}" unless rnext =~ /^(\*|=|[!-()+-<>-~][!-~]*)$/
464 unless @header.empty? or rnext == '*' or rnext == '='
465 unless @header[:SQ][:SN][rnext.to_sym]
466 raise SamError, "rnext not found in header hash: #{rnext}"
471 # Method to check pnext.
472 def check_pnext(pnext)
473 raise SamError, "Bad pnext: #{pnext}" unless (0 .. 2**29 - 1).include? pnext
476 # Method to check tlen.
478 raise SamError, "Bad tlen: #{tlen}" unless (-2**29 + 1 .. 2**29 - 1).include? tlen
481 # Method to check seq.
483 raise SamError, "Bad seq: #{seq}" unless seq =~ /^(\*|[A-Za-z=.]+)$/
486 # Method to check qual.
488 raise SamError, "Bad qual: #{qual}" unless qual =~ /^[!-~]+$/
491 # Method to deconvolute the SAM flag field.
495 # Method to initialize a Flag object.
500 # Method to test if template have
501 # multiple fragments in sequencing.
503 (flag & FLAG_MULTI) == 0
506 # Method to test if each fragment
507 # properly aligned according to the aligner.
509 (flag & FLAG_ALIGNED) == 0
512 # Method to test if the fragment was unmapped.
514 (flag & FLAG_UNMAPPED) == 0
517 # Method to test if the next fragment was unmapped.
519 (flag & FLAG_NEXT_UNMAPPED) == 0
522 # Method to test if the fragment was reverse complemented.
524 (flag & FLAG_REVCOMP) == 0
527 # Method to test if the next fragment was reverse complemented.
529 (flag & FLAG_NEXT_REVCOMP) == 0
532 # Method to test if the fragment was first in the template.
534 (flag & FLAG_FIRST) == 0
537 # Method to test if the fragment was last in the template.
539 (flag & FLAG_LAST) == 0
542 # Method to test for secondary alignment.
543 def secondary_alignment?
544 (flag & FLAG_SECONDARY_ALIGNMENT) == 0
547 # Method to test for quality fail.
549 (flag & FLAG_QUALITY_FAIL) == 0
552 # Method to test for PCR or optical duplicates.
554 (flag & FLAG_DUPLICATES) == 0
560 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<