#sdm options file to control sequence quality filtering, demultiplexing and preparation (can also be used without demultiplexing)
#* indicates alternative quality filtering options, saved in *.add.fna etc. files separately from initial quality filtered dataset
#sequence length refers to sequence length AFTER removal of Primers, Barcodes and trimming. this ensures that downstream analyis tools will have appropiate sequence information
minSeqLength	90
maxSeqLength	200
minAvgQuality	25
*minSeqLength	80
*minAvgQuality	18
#truncate total Sequence length to X (length after Barcode, Adapter and Primer removals, set to -1 to deactivate)
TruncateSequenceLength	90

#Ambiguous bases in Sequence
maxAmbiguousNT	0
*maxAmbiguousNT	1

#sequence is discarded if a homonucleotide run in sequence is longer
maxHomonucleotide	12

#Filter whole sequence if one window of quality scores is below average
QualWindowWidth	50
QualWindowThreshhold	25

#Trim the end of a sequence if a window falls below quality threshhold. Useful for removing low qulaity trailing ends of sequence
TrimWindowWidth	20
TrimWindowThreshhold	0

#Probabilistic max number of accumulated sequencing errors. After this length, the rest of the sequence will be deleted. Complimentary to TrimWindowThreshhold. (-1) deactivates this option.
maxAccumulatedError	0.5
#mid qual option
*maxAccumulatedError	-1
#Binomial error model of expected errors per sequence (see https://github.com/fpusan/moira), to deactivate, set BinErrorModelAlpha to -1
BinErrorModelMaxExpError	2.5
BinErrorModelAlpha	0.005


#Max Barcode Errors
maxBarcodeErrs	0
maxPrimerErrs	0

#keep Barcode / Primer Sequence in the output fasta file - in a normal 16S analysis this should be deactivated (0) for Barcode and de-activated (0) for primer
keepBarcodeSeq	0
keepPrimerSeq	0


#set fastqVersion to 1 if you use Sanger, Illumina 1.8+ or NCBI SRA files. Set fastqVersion to 2, if you use Illumina 1.3+ - 1.7+ or Solexa fastq files. "auto" will look for typical characteristics of either of these and choose the quality offset score automatically.
fastqVersion	auto

#if one or more files have a technical adapter still included (e.g. TCAG 454) this can be removed by setting this option
TechnicalAdapter	

#delete X NTs (e.g. if the first 5 bases are known to have strange biases)
TrimStartNTs	0

#correct PE header format (1/2) this is to accomodate the illumina miSeq paired end annotations 2="@XXX 1:0:4" insteand of 1="@XXX/1". Note that the format will be automatically detected
PEheaderPairFmt	1

#sets if sequences without match to reverse primer (ReversePrimer) will be accepted (T=reject ; F=accept all); default=F
RejectSeqWithoutRevPrim	F
*RejectSeqWithoutRevPrim	F
#sets if sequences without a forward (LinkerPrimerSequence) primer will be accepted (T=reject ; F=accept all); default=F
RejectSeqWithoutFwdPrim	F
*RejectSeqWithoutFwdPrim	F

#this option should be "T" if your amplicons are possibly shorter than a single read in a paired end sequencing run (e.g. if the 16S amplicon length is 200bp in a 250x2 miSeq run, set this to "T"). This option increases runtime by 10%, if in doubt just set to "T". *Requires* LinkerPrimerSequence and ReversePrimer to be defined in mapping file.
AmpliconShortPE	T

#options for difficulties during sequencing library construction
#checks if pair1 and pair2 were switched (ignore if single read data)
CheckForMixedPairs	T
#checks if whole amplicon was reverse-transcribed sequenced (not switched, just reverse translated)
CheckForReversedSeqs	T
