Related Experiment Video

Updated: Aug 6, 2026

Whole Genome Sequencing of Candida glabrata for Detection of Markers of Antifungal Drug Resistance
08:45

Whole Genome Sequencing of Candida glabrata for Detection of Markers of Antifungal Drug Resistance

Published on: December 28, 2017

Genomic variant-driven prediction of azole resistance in Aspergillus fumigatus using GWAS and machine learning

Dingchen Li1,2, Xinkai Yue1,3,4, Wenjuan Hu1,4

  • 1Department for Disinfection and Infection Control, Chinese PLA Center for Disease Control and Prevention, Beijing, China.

Frontiers in Microbiology
|July 23, 2026
PubMed

Insights

Machine learning identified novel azole resistance loci in Aspergillus fumigatus beyond known mutations, improving antifungal surveillance and treatment strategies. This approach enhances understanding of drug resistance in invasive aspergillosis.

Area of Science:

  • Medical Mycology and Infectious Diseases
  • Computational Biology and Bioinformatics
  • Antimicrobial Resistance Research

Background:

  • Azole resistance in Aspergillus fumigatus, a primary cause of invasive aspergillosis, poses a significant public health threat.
  • Existing research primarily focuses on clinical isolates and known genetic drivers like cyp51A/B mutations, leaving gaps in understanding broader genomic contributions to resistance.

Purpose of the Study:

  • To develop a machine learning (ML) framework to identify novel genetic loci associated with azole resistance in Aspergillus fumigatus, moving beyond traditional Genome-Wide Association Studies (GWAS).
  • To analyze a global collection of strains to uncover new resistance mechanisms and inform antifungal strategies.

Main Methods:

  • Whole-genome sequencing (WGS) and resistance phenotyping of 590 global Aspergillus fumigatus strains.
  • Phylogenetic analysis to understand strain diversity and resistance distribution across clades.
  • Genome-Wide Association Studies (GWAS) followed by ML modeling (Random Forest, XGBoost, penalized logistic regression) using high-confidence SNPs for resistance prediction.

Main Results:

  • Phylogenetic analysis revealed four clades; Clade III showed the highest azole resistance rates, primarily linked to cyp51A tandem repeats.
  • ML models, particularly Random Forest and XGBoost, achieved high accuracy (>88%) and AUC (>95%) in predicting resistance.
  • A novel SNP (SCM000172.1_1781459) was identified as a key predictor for resistance to itraconazole, posaconazole, and voriconazole.
  • Cross-resistance analysis indicated overlapping loci for itraconazole and posaconazole resistance, but distinct loci for voriconazole resistance.

Conclusions:

  • The ML framework effectively identifies novel genetic loci contributing to azole resistance in Aspergillus fumigatus.
  • Findings highlight distinct and overlapping resistance mechanisms for different azoles, informing targeted surveillance and drug development.
  • This study provides actionable insights for improving antifungal resistance monitoring and developing tailored treatment strategies for invasive aspergillosis.
Keywords:
Aspergillus fumigatusGWASSNP callingazole resistancemachine learning

More Related Videos

Isolation of Culturable Yeasts and Molds from Soils to Investigate Fungal Population Structure10:33

Isolation of Culturable Yeasts and Molds from Soils to Investigate Fungal Population Structure

Published on: May 27, 2022

Infection of Zebrafish Larvae with Aspergillus Spores for Analysis of Host-Pathogen Interactions09:42

Infection of Zebrafish Larvae with Aspergillus Spores for Analysis of Host-Pathogen Interactions

Published on: May 16, 2020

Related Concept Videos

Genome-wide Association Studies-GWAS01:11

Genome-wide Association Studies-GWAS

Genome-wide association studies or GWAS are used to identify whether common SNPs are associated with certain diseases. Suppose specific SNPs are more frequently observed in individuals with a particular disease than those without the disease. In that case, those SNPs are said to be associated with the disease. Chi-square analysis is performed to check the probability of the allele likely to be associated with the disease.
GWAS does not require the identification of the target gene involved in...
Fungal Phylum Ascomycota01:28

Fungal Phylum Ascomycota

Phylum Ascomycota, a major division within the subkingdom Dikarya, comprises a diverse range of fungal species, including both unicellular yeasts and filamentous molds such as Aspergillus and Penicillium. These fungi thrive in a variety of habitats, from aquatic ecosystems to terrestrial environments, playing crucial ecological and economic roles.Morphology and ReproductionThe defining characteristic of Ascomycetes, commonly referred to as sac fungi, is the ascus—a sac-like structure that...
Pharmacogenomics: Identification of New Drug Targets01:29

Pharmacogenomics: Identification of New Drug Targets

Advances in genomics have profoundly influenced drug discovery by increasing both the speed and accuracy of pharmaceutical development. Pharmacogenomics, which examines how genetic variation influences drug response, facilitates the identification of novel therapeutic targets and enables patient stratification for personalized treatment. These strategies contribute to improved drug efficacy, minimized adverse effects, and more efficient clinical trial design.Mapping genetic differences...