add Otu reference sequence table

[mdoering]

to hold information about reference sequence incl cluster method metadata for OTU based taxa.

[thomasstjerne]

Starting an attribute list here. Should all metadata about clustering method etc be repeated on each OTU?
There are conceptually two entities, the cluster and the chosen representative sequence, but for now these are flattened in one entity.

FieldName	Description	Example data
ID		BOLD:ACG2675, SH1519649.08FU
doi	Doi for the OTU	10.15156/BIO/SH215668.07FU
marker	The Marker (region) of the DNA	ITS2, COI-5P
referenceSequence	A sequence choosen to represent the cluster (OTU)	AACACTTTATTTTATTTTTGGAATTTGATCTGGAATAAT......
genbankAccessionNumber	Link to the ref seq in GenBank	DQ350842
typeStatus	If the rep seq was taken from a type	HOLOTYPE, ISOTYPE
memberCount	Number of sequences in this cluster	5
averageInternalDistance	The average distance between members of the OTU	0.94
maximumInternalDistance	The max distance between members of the OTU	1.79
distanceToNearestNeighbor	Distance to nearest neighbor OTU	9.12
clusteringMethodDesciption	A short description of the clustering method. (Tree-based, Threshold-based, Ref Seq selection, centroid)	OTU Designation Through Refined Single Linkage (RESL) Analysis

[mdoering]

If feasable I think we should put all fields into one flattened entity and repeat metadata - it is not safe to assume the entire dataset shares them. We might surely mix some in the catalogue and potentially derived "downloads". The clustering method description is obviously not nice.

As for ID and DOI I guess that is already part of the taxon entity that this links to? but it needs a taxonID to refer to the taxon