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

Bugs in the non stric tag matching algorithm

Browse Source 
Former-commit-id: efad98d6521322209ce22bc58fab54c7bddd19ec
This commit is contained in:
Eric Coissac
2024-06-10 21:19:26 +02:00
parent 2d4813614b
commit 44533fb61e
3 changed files with 25 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

45
pkg/obingslibrary/multimatch.go
View File

@ -146,11 +146,10 @@ func (marker *Marker) beginFixedTagExtractor(
fb := begin - spacer - taglength fb := begin - spacer - taglength
if fb < 0 { if fb < 0 {
log.Warnf("begin %v", fb)
return "" return ""
} }
return sequence.String()[fb : begin-spacer] return sequence.String()[fb:(begin - spacer)]
} }
func (marker *Marker) endDelimitedTagExtractor( func (marker *Marker) endDelimitedTagExtractor(
@ -266,11 +265,11 @@ func (library *NGSLibrary) TagExtractor(
reverse_tag := marker.endTagExtractor(sequence, end, forward) reverse_tag := marker.endTagExtractor(sequence, end, forward)
if forward_tag != "" { if forward_tag != "" {
annotations["forward_tag"] = forward_tag annotations["obimultiplex_forward_tag"] = forward_tag
} }
if reverse_tag != "" { if reverse_tag != "" {
annotations["reverse_tag"] = reverse_tag annotations["obimultiplex_reverse_tag"] = reverse_tag
} }
return &TagPair{forward_tag, reverse_tag} return &TagPair{forward_tag, reverse_tag}
@ -286,7 +285,7 @@ func (marker *Marker) ClosestForwardTag(
for ts := range marker.samples { for ts := range marker.samples {
d := dist(ts.Forward, tag) d := dist(ts.Forward, tag)
if d == mindist { if d == mindist && mintag != "" && ts.Forward != mintag {
mintag = "" mintag = ""
} }
@ -309,7 +308,7 @@ func (marker *Marker) ClosestReverseTag(
for ts := range marker.samples { for ts := range marker.samples {
d := dist(ts.Reverse, tag) d := dist(ts.Reverse, tag)
if d == mindist { if d == mindist && mintag != "" && ts.Reverse != mintag {
mintag = "" mintag = ""
} }
@ -382,7 +381,7 @@ func (library *NGSLibrary) SampleIdentifier(
pcr, ok := marker.samples[proposed] pcr, ok := marker.samples[proposed]
if !ok { if !ok {
annotations["demultiplex_error"] = fmt.Sprintf("Cannot associate sample to the tag pair (%s:%s)", forward, reverse) annotations["obimultiplex_error"] = fmt.Sprintf("Cannot associate sample to the tag pair (%s:%s)", forward, reverse)
return nil return nil
} }
@ -489,47 +488,47 @@ func (library *NGSLibrary) ExtractMultiBarcode(sequence *obiseq.BioSequence) (ob
if match.Marker == -from.Marker && match.Forward == from.Forward { if match.Marker == -from.Marker && match.Forward == from.Forward {
barcode_error := false barcode_error := false
annotations := obiseq.GetAnnotation() annotations := obiseq.GetAnnotation()
annotations["forward_primer"] = primerseqs[from.Marker].Forward annotations["obimultiplex_forward_primer"] = primerseqs[from.Marker].Forward
annotations["reverse_primer"] = primerseqs[from.Marker].Reverse annotations["obimultiplex_reverse_primer"] = primerseqs[from.Marker].Reverse
if from.Forward { if from.Forward {
if from.Begin < 0 || from.End > sequence.Len() { if from.Begin < 0 || from.End > sequence.Len() {
barcode_error = true barcode_error = true
annotations["multiplex_error"] = "Cannot extract forward match" annotations["obimultiplex_error"] = "Cannot extract forward match"
} else { } else {
annotations["forward_match"] = sequence.String()[from.Begin:from.End] annotations["obimultiplex_forward_match"] = sequence.String()[from.Begin:from.End]
} }
sseq, err := sequence.Subsequence(match.Begin, match.End, false) sseq, err := sequence.Subsequence(match.Begin, match.End, false)
if err != nil { if err != nil {
barcode_error = true barcode_error = true
annotations["multiplex_error"] = "Cannot extract reverse match" annotations["obimultiplex_error"] = "Cannot extract reverse match"
} else { } else {
annotations["reverse_match"] = sseq.ReverseComplement(true).String() annotations["obimultiplex_reverse_match"] = sseq.ReverseComplement(true).String()
} }
annotations["forward_error"] = from.Mismatches annotations["obimultiplex_forward_error"] = from.Mismatches
annotations["reverse_error"] = match.Mismatches annotations["obimultiplex_reverse_error"] = match.Mismatches
} else { } else {
if from.Begin < 0 || from.End > sequence.Len() { if from.Begin < 0 || from.End > sequence.Len() {
barcode_error = true barcode_error = true
annotations["multiplex_error"] = "Cannot extract reverse match" annotations["obimultiplex_error"] = "Cannot extract reverse match"
} else { } else {
annotations["reverse_match"] = sequence.String()[from.Begin:from.End] annotations["obimultiplex_reverse_match"] = sequence.String()[from.Begin:from.End]
} }
sseq, err := sequence.Subsequence(match.Begin, match.End, false) sseq, err := sequence.Subsequence(match.Begin, match.End, false)
if err != nil { if err != nil {
barcode_error = true barcode_error = true
annotations["multiplex_error"] = "Cannot extract forward match" annotations["obimultiplex_error"] = "Cannot extract forward match"
} else { } else {
annotations["forward_match"] = sseq.ReverseComplement(true).String() annotations["obimultiplex_forward_match"] = sseq.ReverseComplement(true).String()
} }
annotations["reverse_error"] = from.Mismatches annotations["obimultiplex_reverse_error"] = from.Mismatches
annotations["forward_error"] = match.Mismatches annotations["obimultiplex_forward_error"] = match.Mismatches
} }
if !barcode_error { if !barcode_error {
@ -572,11 +571,11 @@ func (library *NGSLibrary) ExtractMultiBarcode(sequence *obiseq.BioSequence) (ob
} }
if len(results) == 0 { if len(results) == 0 {
sequence.SetAttribute("demultiplex_error", "No barcode identified") sequence.SetAttribute("obimultiplex_error", "No barcode identified")
results = append(results, sequence) results = append(results, sequence)
} else { } else {
for i, result := range results { for i, result := range results {
result.SetAttribute("amplicon_rank", fmt.Sprintf("%d/%d", i+1, len(results))) result.SetAttribute("obimultiplex_amplicon_rank", fmt.Sprintf("%d/%d", i+1, len(results)))
} }
} }

2
pkg/obioptions/version.go
View File

@ -7,7 +7,7 @@ import (
// TODO: The version number is extracted from git. This induces that the version // TODO: The version number is extracted from git. This induces that the version
// corresponds to the last commit, and not the one when the file will be // corresponds to the last commit, and not the one when the file will be
// commited // commited
var _Commit = "748a235" var _Commit = "5611fb1"
var _Version = "Release 4.2.0" var _Version = "Release 4.2.0"
// Version returns the version of the obitools package. // Version returns the version of the obitools package.

4
pkg/obitools/obimultiplex/demultiplex.go
View File

@ -35,13 +35,13 @@ func IExtractBarcode(iterator obiiter.IBioSequence) (obiiter.IBioSequence, error
if !CLIConservedErrors() { if !CLIConservedErrors() {
log.Infoln("Discards unassigned sequences") log.Infoln("Discards unassigned sequences")
out = out.FilterOn(obiseq.HasAttribute("demultiplex_error").Not(), obioptions.CLIBatchSize()) out = out.FilterOn(obiseq.HasAttribute("obimultiplex_error").Not(), obioptions.CLIBatchSize())
} }
var unidentified obiiter.IBioSequence var unidentified obiiter.IBioSequence
if CLIUnidentifiedFileName() != "" { if CLIUnidentifiedFileName() != "" {
log.Printf("Unassigned sequences saved in file: %s\n", CLIUnidentifiedFileName()) log.Printf("Unassigned sequences saved in file: %s\n", CLIUnidentifiedFileName())
unidentified, out = newIter.DivideOn(obiseq.HasAttribute("demultiplex_error"), unidentified, out = newIter.DivideOn(obiseq.HasAttribute("obimultiplex_error"),
obioptions.CLIBatchSize()) obioptions.CLIBatchSize())
go func() { go func() {