Assembling and auditing a comprehensive DNA barcode reference library for European marine fishes

A large-scale comprehensive reference library of DNA barcodes for European marine fishes was assembled, allowing the evaluation of taxonomic uncertainties and species genetic diversity that were otherwise hidden in geographically restricted studies. A total of 4118 DNA barcodes were assigned to 358 species generating 366 Barcode Index Numbers (BIN). Initial examination revealed as much as 141 BIN discordances (more than one species in each BIN). After implementing an auditing and five-grade (A-E) annotation protocol, the number of discordant species BINs was reduced to 44 (13% grade E), while concordant species BINs amounted to 271 (78% grades A and B) and 14 other had insufficient data (grade D). Fifteen species displayed comparatively high intraspecific divergences ranging from 2·6 to 18·5% (grade C), which is biologically paramount information to be considered in fish species monitoring and stock assessment. On balance, this compilation contributed to the detection of 59 European fish species probably in need of taxonomic clarification or re-evaluation. The generalized implementation of an auditing and annotation protocol for reference libraries of DNA barcodes is recommended. The authors would like to thank the three independent reviewers for their helpful reviews of this manuscript. This work was supported by national funds through the FCT I.P. and by the ERDF through the COMPETE2020 – Program Operacional Competitividade e Internacionalização (POCI) in the scope of the strategic programme UID/BIA/04050/2013 (POCI-01-0145-FEDER-007569). PS is supported by FCT, ESF, POPH and the FCT Investigator Programme (IF/01641/2013). ML was supported by an FCT fellowship with reference SFRH/BPD/45246/2008. We thank R. Martins and M. Carneiro (IPMA, Portugal), as well as A. Kijewska and R. Wenne (Institute of Oceanology, Polish Academy of Sciences, Poland) for providing specimens from Portugal and from the Baltic Sea, respectively, used to generate the new sequences here reported.