FunGeneClusterS: Predicting fungal gene clusters from genome and transcriptome data

Area of Science:

• Genomics
• Bioinformatics
• Mycology

Background:

• Fungal secondary metabolites are of interest for their bioactivities and biosynthesis via gene clusters.
• Previous prediction methods were limited in user-friendliness and parameter adjustment, requiring manual curation.
• This work aimed to improve the accuracy and usability of fungal gene cluster prediction.

Purpose of the Study:

• To develop an improved computational tool for predicting fungal gene clusters.
• To enhance the accuracy and reduce manual curation in identifying co-regulated gene clusters.
• To explore the genomic organization of both primary and secondary metabolism in fungi.

Main Methods:

• Developed FunGeneClusterS, an improved implementation with a graphical user interface.
• Incorporated adjustable parameters for gene cluster size and flexibility in co-regulation.
• Benchmarked the method using genomic and transcriptomic data from *Aspergillus nidulans* and *A. niger*.

Main Results:

• FunGeneClusterS accurately predicts larger gene clusters (>10 genes) and identifies co-regulated clusters irrespective of gene function.
• The method significantly reduces the need for manual curation of predictions.
• Applied to *A. niger*, FunGeneClusterS identified 432 co-transcribed gene clusters, including those for primary metabolism (e.g., biotin biosynthesis) and aromatic compound degradation.

Conclusions:

• FunGeneClusterS is a versatile tool for predicting co-regulated gene clusters, applicable beyond secondary metabolism.
• The study validates FunGeneClusterS and suggests extensive co-regulation and co-location of fungal genes in primary metabolism and cellular functions.
• The findings indicate that larger portions of the fungal genome than previously thought operate as gene clusters.