In order to establish DNA barcoding system for identifying herbal medicine, professor Shilin Chen’s group has obtained foundation supported by International S&T Cooperation Project and National High-tech R&D Program (863 Program). Prof.Chen’s group has selected and compared different candidate barcodes from nuclear, mitochondria and plastid genomes for near ten years. 
The DNA barcoding system for identifying herbal materials has been established based on the ITS2 and psbA–trnH barcodes. The system has been used for identifying herbal materials, and the principles and methods have been approved for incorporation into supplement 3 of the Chinese Pharmacopoeia (2010 edition). Furthermore, an online DNA barcoding database for herbal materials has been constructed (http://www.tcmbarcode.cn), which provides a species identification module for herbal materials. In this database, ITS2 and psbA–trnH are selected as the core and supplementary DNA barcodes, respectively, for herbal materials. In order to correctly distinguish medicinal species, this database also contains barcoding data for their adulterants, substitutes and closely related species. The database includes almost all of materia medica listed in Chinese Pharmacopoeia, Japanese Pharmacopoeia, Korean Pharmacopoeia, Indian Pharmacopeia, United States Pharmacopeia, and European Pharmacopoeia. The data in this database will be updated every six months. 
The data includes, besides sequence information, some other valuable information on voucher specimens and corresponding references. The method for DNA barcode of herbal materials is also included in this database, which provides a standard workflow and guidelines on how to make species identification using DNA barcoding step by step for beginners. The database can be freely accessed by any users who interested in DNA barcode of herbal materials. Users can query their sequences using identification module by pasting them into the input box for a specified identification function. These query sequences should be longer than 150 bp and can be fasta format with “>” identifier or just nucleotides without “>” identifier. After clicking the submit button, a BLASTN algorithm is activated and its nearest neighbors from all reference sequences are made available. For the species included in the database, the identification module can give a species level identification where a tight match to a reference sequence has been found. When a species level match is found, the Latin scientific name of best match species will be given and other related data available in the database can also be connected.