Abstract:
Introduction: Molecular divergence thresholds have been proposed to distinguish recently separated evolutive
units, often displaying more accurate putative species assignments in taxonomic research compared to traditional
morphological approaches. This makes DNA barcoding an attractive identification tool for a variety of marine
invertebrates, especially for cryptic species complexes. Although GenBank and the Barcode of Life Data System
(BOLD) are the major sequence repositories worldwide, very few have tested their performance in the identification of echinoderm sequences.
Objective: We use COI echinoderm sequences from local samples and the molecular identification platforms
from GenBank and BOLD, in order to test their accuracy and reliability in the DNA barcoding identification for
Central American shallow water echinoderms, at genus and species level.
Methods: We conducted sampling, tissue extraction, COI amplification, sequencing, and taxonomic identification for 475 specimens. The 348 obtained sequences were individually enquired with BLAST in GenBank as well
as using the Identification System (IDS) in BOLD. Query sequences were classified depending on the best match
result. McNemar’s chi-squared, Kruskal-Wallis’s and Mann-Whitney’s U tests were performed to prove differences between the results from both databases. Additionally, we recorded an updated list of species reported for
the shallow waters of the Central American Pacific.
Results: We found 324 echinoderm species reported for Central American Pacific shallow waters. Only 118 and
110 were present in GenBank and BOLD databases respectively. We proposed 325 solved morphology-based
identities and 21 provisional identifications in 50 putative taxa. GenBank retrieved 348 molecular-based identifications in 58 species, including twelve provisional identifications in tree taxa. BOLD recovered 170 COI identifications in 23 species with one provisional identification. Nevertheless, 178 sequences retrieved unmatched
terms (in 34 morphology-based taxa). Only 86 sequences (25 %) were retrieved as correct identifications and
128 (37 %) as identification errors in both platforms. We include 84 sequences for eleven species not represented
in GenBank and 65 sequences for ten species in BOLD Echinoderm COI databases. The identification accuracy
using BLAST (175 correct and 152 incorrect identifications) was greater than with IDS engine (110 correct and
218 identification errors), therefore GenBank outperforms BOLD (Kruskal-Wallis = 41.625, df = 1, p < 0.001).
Conclusions: Additional echinoderm sample references are needed to improve the utility of the evaluated DNA
barcoding identification tools. Identification discordances in both databases may obey specific parameters used
in each search algorithm engine and the available sequences. We recommend the use of barcoding
