3 # Copyright (C) 2006-2009 Martin A. Hansen.
5 # This program is free software; you can redistribute it and/or
6 # modify it under the terms of the GNU General Public License
7 # as published by the Free Software Foundation; either version 2
8 # of the License, or (at your option) any later version.
10 # This program is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 # GNU General Public License for more details.
15 # You should have received a copy of the GNU General Public License
16 # along with this program; if not, write to the Free Software
17 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
19 # http://www.gnu.org/copyleft/gpl.html
22 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> DESCRIPTION <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
25 # Routines for manipulation of FASTQ files and FASTQ entries.
27 # http://maq.sourceforge.net/fastq.shtml
30 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
37 use vars qw( @ISA @EXPORT );
45 @ISA = qw( Exporter );
47 use Inline ( C => <<'END_C', DIRECTORY => $ENV{ "BP_TMP" } );
50 About Phred or FASTQ format:
52 This format is an adaption of the fasta format that contains quality scores.
53 However, the fastq format is not completely compatible with the fastq files
54 currently in existence, which is read by various applications (for example,
55 BioPerl). Because a larger dynamic range of quality scores is used, the
56 quality scores are encoded in ASCII as 64+score, instead of the standard
57 32+score. This method is used to avoid running into non-printable characters.
60 About Solexa or SCARF format:
62 SCARF (Solexa compact ASCII read format)—This easy-to-parse text based format,
63 stores all the information for a single read in one line. Allowed values are
64 --numeric and --symbolic. --numeric outputs the quality values as a
65 space-separated string of numbers. --symbolic outputs the quality values as a
66 compact string of ASCII characters. Subtract 64 from the ASCII code to get the
67 corresponding quality values. Under the current numbering scheme, quality
68 values can theoretically be as low as -5, which has an ASCII encoding of 59=';'.
72 http://maq.sourceforge.net/fastq.shtml */
75 int phred2dec( char c )
77 /* Martin A. Hansen, July 2009 */
79 /* Converts a Phred score in octal (a char) to a decimal score, */
80 /* which is returned. */
84 score = ( int ) c - 33;
90 int solexa2dec( char c )
92 /* Martin A. Hansen, July 2009 */
94 /* Converts a Solexa score in octal (a char) to a decimal score, */
95 /* which is returned. */
99 score = ( int ) c - 64;
105 // int solexa2dec( char c )
107 // /* Martin A. Hansen, July 2009 */
109 // /* Converts a Solexa score in octal (a char) to a decimal score, */
110 // /* which is returned. */
112 // /* http://maq.sourceforge.net/fastq.shtml */
113 // /* $Q = 10 * log(1 + 10 ** (ord($sq) - 64) / 10.0)) / log(10); */
116 // int ascii = ( int ) c - 64;
118 // score = 10 * log( 1 + pow( 10, ascii / 10 ) ) / log( 10 );
120 // // printf( "char: %c ascii: %d score: %d\n", c, ascii, score );
126 char dec2phred( int score )
128 /* Martin A. Hansen, July 2009 */
130 /* Converts a decimal score to an octal score (a char) in the Phred range, */
131 /* which is returned. */
135 c = ( char ) score + 33;
141 char dec2solexa( int score )
143 /* Martin A. Hansen, September 2009 */
145 /* Converts a decimal score to an octal score (a char) in the Solexa range, */
146 /* which is returned. */
150 c = ( char ) score + 64;
156 char solexa2phred( char score )
158 /* Martin A. Hansen, September 2009 */
160 /* Converts an octal score in the Solexa range to an octal score */
161 /* in the Pred range, which is returned. */
166 dec = solexa2dec( score );
173 char phred2solexa( char score )
175 /* Martin A. Hansen, September 2009 */
177 /* Converts an octal score in the Solexa range to an octal score */
178 /* in the Pred range, which is returned. */
183 dec = phred2dec( score );
190 void solexa_str2phred_str( char *scores )
192 /* Martin A. Hansen, September 2009 */
194 /* Converts a string of Solexa octal scores to Phred octal scores in-line. */
198 for ( i = 0; i < strlen( scores ); i++ ) {
199 scores[ i ] = solexa2phred( scores[ i ] );
204 void phred_str2solexa_str( char *scores )
206 /* Martin A. Hansen, September 2009 */
208 /* Converts a string of Phred octal scores to Solexa octal scores in-line. */
212 for ( i = 0; i < strlen( scores ); i++ ) {
213 scores[ i ] = phred2solexa( scores[ i ] );
218 double phred_str_mean( char *scores )
220 /* Martin A. Hansen, November 2009 */
222 /* Calculates the mean score as a float which is retuned. */
229 len = strlen( scores );
231 for ( i = 0; i < len; i++ ) {
232 sum += phred2dec( scores[ i ] );
235 mean = ( double ) sum / ( double ) len;
241 void softmask_solexa_str( char *seq, char *scores, int threshold )
243 /* Martin A. Hansen, July 2009 */
245 /* Given a sequence string and a score string (in Solexa range) */
246 /* lowercases all residues where the decimal score is below a given threshold. */
250 for ( i = 0; i < strlen( seq ); i++ )
252 if ( solexa2dec( scores[ i ] ) < threshold ) {
253 seq[ i ] = tolower( seq[ i ] );
259 void softmask_phred_str( char *seq, char *scores, int threshold )
261 /* Martin A. Hansen, July 2009 */
263 /* Given a sequence string and a score string (in Phred range) */
264 /* lowercases all residues where the decimal score is below a given threshold. */
268 for ( i = 0; i < strlen( seq ); i++ )
270 if ( phred2dec( scores[ i ] ) < threshold ) {
271 seq[ i ] = tolower( seq[ i ] );
280 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
283 sub solexa_str2dec_str
285 # Martin A. Hansen, September 2009
287 # Converts a string of Solexa octal scores to a ; separated
288 # string of decimal scores.
290 my ( $scores, # Solexa scores
295 $scores =~ s/(.)/ solexa2dec( $1 ) . ";"/eg;
301 sub phred_str2dec_str
303 # Martin A. Hansen, September 2009
305 # Converts a string of Phred octal scores to a ; separated
306 # string of decimal scores.
308 my ( $scores, # Phred scores
313 $scores =~ s/(.)/ phred2dec( $1 ) . ";"/eg;
321 # Martin A. Hansen, July 2009.
323 # Gets the next FASTQ entry from a given filehandle.
325 my ( $fh, # filehandle
330 my ( $seq, $seq_name, $qual, $qual_name );
346 return wantarray ? ( $seq_name, $seq, $qual ) : [ $seq_name, $seq, $qual ];
352 # Martin A. Hansen, July 2009.
354 # Output a FASTQ entry to STDOUT or a filehandle.
356 my ( $entry, # FASTQ entry
357 $fh, # filehandle - OPTIONAL
364 print $fh "@" . $entry->[ SEQ_NAME ] . "\n";
365 print $fh $entry->[ SEQ ] . "\n";
367 print $fh $entry->[ SCORES ] . "\n";
373 # Martin A. Hansen, July 2009.
375 # Converts a FASTQ entry to a Biopiece record, where
376 # the FASTQ quality scores are converted to numerics.
378 my ( $entry, # FASTQ entry,
385 $record->{ 'SEQ' } = $entry->[ SEQ ];
386 $record->{ 'SEQ_NAME' } = $entry->[ SEQ_NAME ];
387 $record->{ 'SCORES' } = $entry->[ SCORES ];
388 $record->{ 'SEQ_LEN' } = length $entry->[ SEQ ];
390 return wantarray ? %{ $record } : $record;
396 # Martin A. Hansen, July 2009.
398 # Converts a Biopiece record to a FASTQ entry.
400 my ( $record, # Biopiece record
407 if ( exists $record->{ 'SEQ' } and exists $record->{ 'SEQ_NAME' } and exists $record->{ 'SCORES' } )
409 $list->[ SEQ_NAME ] = $record->{ 'SEQ_NAME' };
410 $list->[ SEQ ] = $record->{ 'SEQ' };
411 $list->[ SCORES ] = $record->{ 'SCORES' };
413 $list->[ SCORES ] =~ s/(\d+);/chr( ( $1 <= 93 ? $1 : 93 ) + 33 )/ge;
415 return wantarray ? @{ $list } : $list;
422 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<