obitools/obitools4
2
0
Fork 0
mirror of https://github.com/metabarcoding/obitools4.git synced 2025-06-29 16:20:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

optimize sequence readers and patch a bug in the format guesser

Browse Source 
Former-commit-id: 9dce1e96c57ae9a88c26fac5c8e1bdcdc2c0c7a5
This commit is contained in:
Eric Coissac
2023-10-13 21:52:57 +02:00
parent ddf0f2cede
commit e8c55a2b6b
7 changed files with 141 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files

56
pkg/obiapat/pcr.go
View File

@ -14,6 +14,7 @@ type _Options struct {
forwardError int
reverseError int
extension int
fullExtension bool
batchSize int
parallelWorkers int
forward ApatPattern
@ -41,6 +42,10 @@ func (options Options) Extension() int {
return options.pointer.extension
}
func (options Options) OnlyFullExtension() bool {
return options.pointer.fullExtension
}
// MinLength method returns minimum length of
// the searched amplicon (length of the primers
// excluded)
@ -96,6 +101,7 @@ func MakeOptions(setters []WithOption) Options {
forwardError: 0,
reverseError: 0,
extension: -1,
fullExtension: false,
circular: false,
parallelWorkers: 4,
batchSize: 100,
@ -172,6 +178,14 @@ func OptionWithExtension(extension int) WithOption {
return f
}
func OptionOnlyFullExtension(full bool) WithOption {
f := WithOption(func(opt Options) {
opt.pointer.fullExtension = full
})
return f
}
// OptionForwardError sets the number of
// error allowed when matching the forward
// primer.
@ -285,7 +299,25 @@ func _Pcr(seq ApatSequence,
from = fm[1]
to = rm[0]
}
amplicon, _ := seq.pointer.reference.Subsequence(from, to, opt.pointer.circular)
if opt.HasExtension() && !opt.OnlyFullExtension() && !opt.Circular() {
if from < 0 {
from = 0
}
if to > seq.Len() {
to = seq.Len()
}
}
if (opt.HasExtension() && ((from >= 0 && to <= seq.Len()) || opt.Circular())) ||
!opt.HasExtension() {
amplicon, error := seq.pointer.reference.Subsequence(from, to, opt.Circular())
if error != nil {
log.Fatalf("error : %v\n", error)
}
log.Debugf("seq length : %d capacity : %d", amplicon.Len(), cap(amplicon.Sequence()))
annot := amplicon.Annotations()
obiutils.MustFillMap(annot, seq.pointer.reference.Annotations())
@ -310,6 +342,8 @@ func _Pcr(seq ApatSequence,
// log.Debugf("amplicon sequence capacity : %d", cap(amplicon.Sequence()))
results = append(results, amplicon)
}
}
}
}
@ -370,7 +404,24 @@ func _Pcr(seq ApatSequence,
from = fm[1]
to = rm[0]
}
amplicon, _ := seq.pointer.reference.Subsequence(from, to, opt.pointer.circular)
if opt.HasExtension() && !opt.OnlyFullExtension() && !opt.Circular() {
if from < 0 {
from = 0
}
if to > seq.Len() {
to = seq.Len()
}
}
if (opt.HasExtension() && ((from >= 0 && to <= seq.Len()) || opt.Circular())) ||
!opt.HasExtension() {
amplicon, error := seq.pointer.reference.Subsequence(from, to, opt.pointer.circular)
if error != nil {
log.Fatalf("error : %v\n", error)
}
amplicon = amplicon.ReverseComplement(true)
annot := amplicon.Annotations()
@ -401,6 +452,7 @@ func _Pcr(seq ApatSequence,
}
}
}
}
return results
}

8
pkg/obiformats/embl_read.go
View File

@ -192,8 +192,10 @@ func _ReadFlatFileChunk(reader io.Reader, readers chan _FileChunk) {
for err == nil {
// Read from the reader until the buffer is full or the end of the file is reached
for ; err == nil && l < len(buff); l += size {
size, err = reader.Read(buff[l:])
l, err = io.ReadFull(reader, buff)
if err == io.ErrUnexpectedEOF {
err = nil
}
// Create an extended buffer to read from if the end of the last entry is not found in the current buffer
@ -205,7 +207,7 @@ func _ReadFlatFileChunk(reader io.Reader, readers chan _FileChunk) {
// Read from the reader in 1 MB increments until the end of the last entry is found
for end = _EndOfLastEntry(buff); err == nil && end < 0; end = _EndOfLastEntry(extbuff[:size]) {
ic++
size, err = reader.Read(extbuff)
size, err = io.ReadFull(reader, extbuff)
buff = append(buff, extbuff[:size]...)
}

16
pkg/obiformats/fastaseq_read.go
View File

@ -92,7 +92,12 @@ func FastaChunkReader(r io.Reader, size int, cutHead bool) (chan FastxChunk, err
out := make(chan FastxChunk)
buff := make([]byte, size)
n, err := r.Read(buff)
n, err := io.ReadFull(r, buff)
if err == io.ErrUnexpectedEOF {
err = nil
}
if n > 0 && err == nil {
if n < size {
buff = buff[:n]
@ -128,13 +133,20 @@ func FastaChunkReader(r io.Reader, size int, cutHead bool) (chan FastxChunk, err
index: idx,
}
idx++
} else {
size = size * 2
}
buff = slices.Grow(buff[:0], size)[0:size]
n, err = r.Read(buff)
n, err = io.ReadFull(r, buff)
if n < size {
buff = buff[:n]
}
if err == io.ErrUnexpectedEOF {
err = nil
}
// fmt.Printf("n = %d, err = %v\n", n, err)
}

14
pkg/obiformats/fastqseq_read.go
View File

@ -99,7 +99,11 @@ func FastqChunkReader(r io.Reader, size int) (chan FastxChunk, error) {
out := make(chan FastxChunk)
buff := make([]byte, size)
n, err := r.Read(buff)
n, err := io.ReadFull(r, buff)
if err == io.ErrUnexpectedEOF {
err = nil
}
if n > 0 && err == nil {
if n < size {
@ -130,13 +134,19 @@ func FastqChunkReader(r io.Reader, size int) (chan FastxChunk, error) {
index: idx,
}
idx++
} else {
size = size * 2
}
buff = slices.Grow(buff[:0], size)[0:size]
n, err = r.Read(buff)
n, err = io.ReadFull(r, buff)
if n < size {
buff = buff[:n]
}
if err == io.ErrUnexpectedEOF {
err = nil
}
// fmt.Printf("n = %d, err = %v\n", n, err)
}

16
pkg/obiformats/universal_read.go
View File

@ -69,11 +69,17 @@ func OBIMimeTypeGuesser(stream io.Reader) (*mimetype.MIME, io.Reader, error) {
mimetype.Lookup("text/plain").Extend(genbankDetector, "text/genbank", ".seq")
mimetype.Lookup("text/plain").Extend(emblDetector, "text/embl", ".dat")
mimetype.Lookup("application/octet-stream").Extend(fastaDetector, "text/fasta", ".fasta")
mimetype.Lookup("application/octet-stream").Extend(fastqDetector, "text/fastq", ".fastq")
mimetype.Lookup("application/octet-stream").Extend(ecoPCR2Detector, "text/ecopcr2", ".ecopcr")
mimetype.Lookup("application/octet-stream").Extend(genbankDetector, "text/genbank", ".seq")
mimetype.Lookup("application/octet-stream").Extend(emblDetector, "text/embl", ".dat")
// Create a buffer to store the read data
buf := make([]byte, 1024*128)
n, err := stream.Read(buf)
n, err := io.ReadFull(stream, buf)
if err != nil && err != io.EOF {
if err != nil && err != io.ErrUnexpectedEOF {
return nil, nil, err
}
@ -84,7 +90,11 @@ func OBIMimeTypeGuesser(stream io.Reader) (*mimetype.MIME, io.Reader, error) {
}
// Create a new reader based on the read data
newReader := io.MultiReader(bytes.NewReader(buf[:n]), stream)
newReader := io.Reader(bytes.NewReader(buf[:n]))
if err == nil {
newReader = io.MultiReader(newReader, stream)
}
return mimeType, newReader, nil
}

8
pkg/obitools/obipcr/options.go
View File

@ -18,6 +18,7 @@ var _MinimumLength = 0
var _MaximumLength = -1
var _Fragmented = false
var _Delta = -1
var _OnlyFull = false
// PCROptionSet defines every options related to a simulated PCR.
//
@ -58,6 +59,9 @@ func PCROptionSet(options *getoptions.GetOpt) {
options.IntVar(&_Delta, "delta", -1,
options.Alias("D"),
options.Description("Lenght of the sequence fragment to be added to the barcode extremities."))
options.BoolVar(&_OnlyFull, "only-complete-flanking", false,
options.Description("Only fragments with complete flanking sequences are printed."))
}
// OptionSet adds to the basic option set every options declared for
@ -131,3 +135,7 @@ func CLIWithExtension() bool {
func CLIExtension() int {
return _Delta
}
func CLIOnlyFull() bool {
return _OnlyFull
}

1
pkg/obitools/obipcr/pcr.go
View File

@ -24,6 +24,7 @@ func CLIPCR(iterator obiiter.IBioSequence) (obiiter.IBioSequence, error) {
CLIReversePrimer(),
CLIAllowedMismatch(),
),
obiapat.OptionOnlyFullExtension(CLIOnlyFull()),
)
if CLIMinLength() > 0 {