Identification of Mycobacterium abscessus Subspecies by MALDI-TOF Mass Spectrometry and Machine Learning

Mycobacterium abscessus is one of the most common and pathogenic nontuberculous mycobacteria (NTM) isolated in clinical laboratories. It consists of three subspecies: M. abscessus subsp. abscessus, M. abscessus subsp. bolletii, and M. abscessus subsp. massiliense. Due to their different antibiotic susceptibility pattern, a rapid and accurate identification method is necessary for their differentiation. Although matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) has proven useful for NTM identification, the differentiation of M. abscessus subspecies is challenging. In this study, a collection of 325 clinical isolates of M. abscessus was used for MALDI-TOF MS analysis and for the development of machine learning predictive models based on MALDI-TOF MS protein spectra. Overall, using a random forest model with several confidence criteria (samples by triplicate and similarity values >60%), a total of 96.5% of isolates were correctly identified at the subspecies level. Moreover, an improved model with Spanish isolates was able to identify 88.9% of strains collected in other countries. In addition, differences in culture media, colony morphology, and geographic origin of the strains were evaluated, showing that the latter had an impact on the protein spectra. Finally, after studying all protein peaks previously reported for this species, two novel peaks with potential for subspecies differentiation were found. Therefore, machine learning methodology has proven to be a promising approach for rapid and accurate identification of subspecies of M. abscessus using MALDI-TOF MS.

Novel Tetraplex Quantitative PCR Assays for Simultaneous Detection and Identification of Xylella fastidiosa Subspecies in Plant Tissues

Xylella fastidiosa (Xf) is an insect-borne bacterium confined to the xylem vessels of plants. This plant pathogen has a broad host range estimated to 560 plant species. Five subspecies of the pathogen with different but overlapping host ranges have been described, but only three subspecies are widely accepted, namely subspecies fastidiosa, multiplex, and pauca. Initially limited to the Americas, Xf has been detected in Europe since 2013. As management of X. fastidiosa outbreaks in Europe depends on the identification of the subspecies, accurate determination of the subspecies in infected plants as early as possible is of major interest. Thus, we developed various tetraplex and triplex quantitative PCR (qPCR) assays for X. fastidiosa detection and subspecies identification in planta in a single reaction. We designed primers and probes using SkIf, a bioinformatics tool based on k-mers, to detect specific signatures of the species and subspecies from a data set of 58 genome sequences representative of X. fastidiosa diversity. We tested the qPCR assays on 39 target and 30 non-target strains, as well as on 13 different plant species spiked with strains of the different subspecies of X. fastidiosa, and on samples from various environmental and inoculated host plants. Sensitivity of simplex assays was equal or slightly better than the reference protocol on purified DNA. Tetraplex qPCR assays had the same sensitivity than the reference protocol and allowed X. fastidiosa detection in all spiked matrices up to 103 cells.ml-1. Moreover, mix infections of two to three subspecies could be detected in the same sample with tetraplex assays. In environmental plant samples, the tetraplex qPCR assays allowed subspecies identification when the current method based on multilocus sequence typing failed. The qPCR assays described here are robust and modular tools that are efficient for differentiating X. fastidiosa subspecies directly in plant samples.

Identification of subspecies-divergent genetic loci responsible for mineral accumulation in rice grains

Rice (Oryza sativa L.) is a major staple food that provides not only dietary calories but also trace elements for the global inhabitants. The insufficiency of mineral nutrients and the potential accumulation of excessive toxic elements in grains pose risks to human health. The substantial natural variations in mineral accumulation in rice grains presents potentials for genetic improvements of rice via biofortifications of essential mineral nutrients and eliminations of toxic elements in grains. However, the genetic mechanisms underlying the natural variations in mineral accumulation have not been fully explored to date owing to unstable phenotypic variations, which are attributed to poor genetic performance and strong environmental effects. In this study, we first compared the genetic performance of different normalization approaches in determining the grain-Cd, grain-Mn, and grain-Zn variations in rice in different genetic populations. Then through quantitative trait loci (QTLs) identification in two rice inter-ectype populations, three QTLs, including qCd7, qMn3, and qZn7, were identified and the QTLs were found to exhibit allelic differentiation in the different ecotypes. Our results were expected to broaden our understanding for mineral accumulation in rice and propose the potential functional alleles that can be explored for further genetic improvement of rice.

Genome-based reclassification of Sphingobacterium soli Fu et al. 2017 as a later heterotypic synonym of Sphingobacterium cellulitidis Huys et al. 2017 and proposal of Sphingobacterium siyangense subsp. siyangense subsp. nov. and Sphingobacterium siyangense subsp. cladoniae subsp. nov

Sphingobacterium, as the type genus of the family Sphingobacteriaceae, comprises a diverse array of species found in various environments. In this study, we aim to reassess and elucidate the taxonomic relationships of Sphingobacterium species. Based on 16S rRNA gene sequences, the phylogeny of 70 validly published Sphingobacterium species was reconstructed. Of which, 50 species with available genomes were further subjected to overall genome relatedness indices (OGRI) analysis, resulting in the identification of distinct pairs of closely related species. One such pair, consisting of the type strains of Sphingobacterium soli and Sphingobacterium cellulitidis, exhibited an average nucleotide identity (ANI) of 97.7%, a digital DNA-DNA hybridization (dDDH) of 80.1% and an average amino acid identity (AAI) of 98.3%, alongside a 16S rRNA gene sequence similarity of 99.8%. Based on the phylogenetic, OGRI and phenotypical evidence, we propose S. soli as a later heterotypic synonym of S. cellulitidis. Additionally, another pair of type strains, Sphingobacterium siyangense and Sphingobacterium cladoniae, possessed ANI, dDDH, AAI and 16S rRNA gene sequence similarity values of 96.3, 70.1, 96.0 and 99.0%, respectively. These values, together with differences in phenotypic traits, support the proposal of two subspecies within this taxonomic lineage. Thus, we propose the establishment of two new subspecies, S. siyangense subsp. siyangense subsp. nov. and S. siyangense subsp. cladoniae subsp. nov.