‘EMALiMB’; A Comprehensive Database for Rapid Identification of Clinical Fungi by MALDI-TOF MS

A collaborative effort among the three microbe culture collections (GCMR, IFM, JCM) has resulted in the development and release of “EMALiMB”, the extended database for rapid identification using MALDI-TOF MS. This system provides a cost-effective and time-saving alternative to sequencing for identification in clinical laboratories, as well as in food and industrial hygiene management. However, fungal identification presented technical challenges and a lack of references compared to bacteria. To address this, a new reference library was constructed, encompassing 74 genera in Ascomycota (e.g. Candida, Starmerella, Debaryomyces, Pichia, Aspergillus), 80 genera in Basidiomycota (e.g. Rhodotorula, Trichosporon, Apiotrichum, Trametes), and Prototheca in Chlorophyta, totaling 633 species and 1,709 main spectral profiles (MSPs). This library aimed to supplement groups with limited strains or lacking registration in the conventional library. More diverse strains, including non-pathogenic species allowed for complementing the genetic factors reflecting Asian locality and improving accuracy. Fungal samples required specific cultivation conditions and preprocessing methods to obtain satisfactory spectra. For example, Malassezia, Basidiomycetous yeasts, showed better MSPs with bead crashing cells from younger mycelia. An additional dataset from 329 clinical isolates were used for validation, comparing answers between the conventional and the expanded libraries. The EMALiMB significantly improved correct answer rates from 84.16% to 93.74% and increased the average score from 2.09 to 2.31. Overall, the ability to identify minor species; such as Candida non-albicans, had reduced cases of unidentifiable. EMALiMB ver. 2 (including bacteria) has been released, and it will continue to be updated in our culture collections by daily operations.