Fix: Allow non-PHRED quality scores e.g. Solexa

Solexa FASTQ reads are not PHRED scores, and as such can be negative.
Relax some of the checks regarding qualities.

Author: jakobnissen

CHANGELOG.md

@@ -4,6 +4,14 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.html).
 
+## [2.1.2]
+### Bugfix
+* Allow non-PHRED quality scores, such as Solexa scores, which can be negative (#104)
+
+## [2.1.0]
+### Bugfix
+* Fix doc examples for writer with do-syntax (#100)
+
 ## [2.1.0]
 ### Additions
 * Implement `Base.copy!` for `FASTQRecord` and `FASTARecord`

Project.toml

@@ -1,7 +1,7 @@
 name = "FASTX"
 uuid = "c2308a5c-f048-11e8-3e8a-31650f418d12"
 authors = ["Sabrina J. Ward <sabrinajward@protonmail.com>", "Jakob N. Nissen <jakobnybonissen@gmail.com>"]
-version = "2.1.1"
+version = "2.1.2"
 
 [weakdeps]
 BioSequences = "7e6ae17a-c86d-528c-b3b9-7f778a29fe59"

src/fastq/quality.jl

@@ -9,11 +9,12 @@
 """
     QualityEncoding(range::StepRange{Char}, offset::Integer)
 
-FASTQ PHRED quality encoding scheme. `QualityEncoding` objects are used to
+FASTQ quality encoding scheme. `QualityEncoding` objects are used to
 interpret the quality scores of FASTQ records.
 `range` is a range of allowed ASCII chars in the encoding, e.g. `'!':'~'` for
 the most common encoding scheme.
-The offset is the PHRED offset.
+The offset is the ASCII offset, i.e. a character with ASCII value `x` encodes
+the value `x - offset`.
 
 See also: [`quality_scores`](@ref)
 
@@ -44,9 +45,7 @@ struct QualityEncoding
         elseif high > 127
             error("Quality encoding only works with ASCII charsets")
         elseif offset < 0
-            error("Quality offset must be non-negative")    
-        elseif low < offset
-            error("Low end of in quality encoding range cannot be less than offset")
+            error("Quality offset must be non-negative")
         else
             return new(low, high, off)
         end
@@ -57,7 +56,7 @@ end
 const SANGER_QUAL_ENCODING     = QualityEncoding('!':'~', 33)
 
 "Solexa (Solexa+64) quality score encoding"
-const SOLEXA_QUAL_ENCODING     = QualityEncoding('@':'~', 64)
+const SOLEXA_QUAL_ENCODING     = QualityEncoding(';':'~', 64)
 
 "Illumina 1.3 (Phred+64) quality score encoding"
 const ILLUMINA13_QUAL_ENCODING = QualityEncoding('@':'~', 64)

test/fastq/TestFASTQ.jl

@@ -4,6 +4,7 @@ module TestFASTQ
 const OFFSET = 33
 
 using ReTest
+using FASTX: FASTQ
 using FASTX.FASTQ: Record, Reader, Writer, identifier, description,
     sequence, quality, quality_scores, QualityEncoding, quality_header!, validate_fastq
 using BioSequences: LongDNA, LongRNA, LongAA, @dna_str, @rna_str, @aa_str
@@ -41,4 +42,4 @@ include("record.jl")
 include("io.jl")
 include("specimens.jl")
 
-end # module TestFASTQ
+end # module TestFASTQ

test/fastq/record.jl

@@ -191,7 +191,6 @@ end
 
     # QualityEncoding
     @test_throws Exception QualityEncoding('B':'A', 10)
-    @test_throws Exception QualityEncoding('A':'A', 90)
     @test_throws Exception QualityEncoding('a':'A', 10)
     @test_throws Exception QualityEncoding('Z':'Y', 10)
     @test_throws Exception QualityEncoding('A':'B', -1)
@@ -208,6 +207,10 @@ end
     @test_throws BoundsError quality_scores(records[2], 2:5)
     @test_throws BoundsError quality_scores(records[2], 5:5)
 
+    # Solexa encoding is weird in thay it can be negative
+    rec = Record("abc", "TAG", [20, 0, -5]; offset=64)
+    @test collect(quality_scores(rec, FASTQ.SOLEXA_QUAL_ENCODING)) == [20, 0, -5]
+
     # Custom quality encoding
     CustomQE = QualityEncoding('A':'Z', 12)
     good = parse(Record, "@a\naaaaaa\n+\nAKPZJO")
@@ -285,4 +288,4 @@ end
     @test isequal(cp, records[2])
 end
 
-end # testset Record
+end # testset Record