Format: ORAL

Ping Wu1, 2

1 Sichuan Normal University, Chengdu, Sichuan, China, 610101 2 Institution of Botany, Chinese Academy of Sciences, Beijing, China, 100093

DNA barcoding, which identifies species using sequences from universal genetic markers, has become a routine method over the past two decades. As for botany, standard DNA barcodes such as rbcL, matK, ITS, and trnL-F have been established, along with some lineage-specific barcodes. Traditional identification methods primarily depend on calculating the genetic distance between the target sequence and those in a database to determine the best match. This approach often overlooks the evolutionary relationships between sequences, such as lineage-specific sequence mutation events and ignores the differential treatment of mutations in conserved versus highly variable regions. Moreover, when the genetic distance between the target sequence and multiple sequences in the database is similar, it can lead to incorrect species identification. Phylogenetic trees can make more comprehensive use of multi-sequence alignment data to construct evolutionary relationships and are theoretically applicable for more accurate species identification. However, building phylogenetic trees requires substantial computational resources, and there is a current lack of automated analysis tools, limiting its use to manual identification for a small number of species. This study has developed PhyloBarcode, a software that automates species identification by constructing and analyzing phylogenetic trees. By preprocessing common barcode sequences, building trees, and optimizing algorithms, this method not only surpasses the accuracy and utility of results compared to traditional methods using genetic distances but also does not significantly lag in running speed. For lineage-specific barcodes, we provide a quick and convenient process for importing data, building reference phylogenetic trees, and matching target sequences. To facilitate ease of use, we also offer an online platform for direct identification, thereby avoiding any technical burden. This research is poised to further enhance the accuracy of plant species identification, bringing convenience to related research and applications.