Automated Identification of Medically Important Bacteria by 16S rRNA Gene Sequencing Using a Novel Comprehensive Database, 16SpathDB

Despite the increasing use of 16S rRNA gene sequencing, interpretation of 16S rRNA gene sequence results is one of the most difficult problems faced by clinical microbiologists and technicians. To overcome the problems we encountered in the existing databases during 16S rRNA gene sequence interpretation, we built a comprehensive database, 16SpathDB (http://147.8.74.24/16SpathDB) based on the 16S rRNA gene sequences of all medically important bacteria listed in the Manual of Clinical Microbiology and evaluated its use for automated identification of these bacteria. Among 91 nonduplicated bacterial isolates collected in our clinical microbiology laboratory, 71 (78%) were reported by 16SpathDB as a single bacterial species having >98.0% nucleotide identity with the query sequence, 19 (20.9%) were reported as more than one bacterial species having >98.0% nucleotide identity with the query sequence, and 1 (1.1%) was reported as no match. For the 71 bacterial isolates reported as a single bacterial species, all results were identical to their true identities as determined by a polyphasic approach. For the 19 bacterial isolates reported as more than one bacterial species, all results contained their true identities as determined by a polyphasic approach and all of them had their true identities as the "best match in 16SpathDB." For the isolate (Gordonibacter pamelaeae) reported as no match, the bacterium has never been reported to be associated with human disease and was not included in the Manual of Clinical Microbiology. 16SpathDB is an automated, user-friendly, efficient, accurate, and regularly updated database for 16S rRNA gene sequence interpretation in clinical microbiology laboratories.