Genomics and transcriptomics of fusarium musae from human and banana

Fusarium musae is a trans-kingdom pathogen that cause crown rot, a post-harvest disease in banana fruits, and systemic and superficial infection in humans. In this work, 18 genomes of F. musae, obtained from human patients and banana fruits, were sequenced using Illumina and MinION platforms. Data were uploaded on the European GALAXY platform, where the following analyses were performed: read quality check and filtering, nuclear (with several tools) and mitochondrial DNA assembly and completeness assessment. One representative strain from each host was manually processed on Geneious Prime to yield chromosome. Final assemblies were then annotated using Funannotate workflow for functional annotation, and then Antismash and EffectorP for secondary metabolite clusters and effectors, respectively. Despite the diversity in F. musae genome size (44 ± 1.5 Mb), strains from bananas and strains from human patients group together in the all the phylogenetic analysis (ML phylogenetic tree, FastANI alignment and MASH comparison) showing no host-effect which suggests that the pathogen is able to adapt to diverse hosts and that transmission from banana to human patients is a possible root of infection. Illumina RNA-Seq data obtained from different media (Blood medium, Banana peel medium, secondary metabolite medium) were mapped and counted using GALAXY pipelines. Transcriptomic data led to the identification of specific under and over regulated genes and secondary metabolites gene clusters showing specialized genomic regions activated specifically in the blood or in the banana media. Moreover, mapped reads improved genome annotation of a representative strain of F. musae. This work is, to our knowledge, the first genomic comparison of a population of F. musae strains obtained from human patients and banana fruits, highlighting possible determinants specific for the infection and adaptation in two different hosts.