ResFungi: A Novel Protein Database of Antifungal Drug Resistance Genes Using a Hidden Markov Model Profile

Abstract

Fungal infections vary from superficial to invasive and can be life-threatening in immunocompromised and healthy individuals. Antifungal resistance is one of the main reasons for an increasing concern about fungal infections as they become more complex and harder to treat. The fungal "omics" databases help us find drug resistance genes, which is of great importance and extremely necessary. With that in mind, we built a new platform for drug resistance genes. We added seven drug classes of resistance genes to our database: azoles (without specifying which drug), fluconazole, voriconazole, itraconazole, flucytosine, micafungin, and caspofungin. Species with known resistance genes were used to validate the results from our database. This study describes a list of 261 candidate genes related to antifungal resistance, with several genes displaying transport functions involved in azole resistance. Over 65% of the candidate genes found were related to at least one type of azole. Overall, the candidate genes found have functional annotations consistent with genes or enzymes that have been linked to antifungal resistance in previous studies. Also, candidate antifungal resistance genes found exhibit functional annotations consistent with previously described resistance mechanisms. The existence of an HMM profile focusing on antifungal resistance genes allows in silico searches for candidate genes, helping future wet lab experiments, and hence, reducing costs when studying candidate antifungal genes without prior knowledge of the species or genes. Finally, ResFungi has proven to be a powerful tool to narrow down candidate antifungal-related genes and unravel mechanisms related to resistance to help in the design of experiments focusing on the genetic basis of antifungal resistance.