Fungen: clustering and correcting long-read metatranscriptomic data for exploring eukaryotic microorganisms.

Area of Science:

• Metagenomics
• Transcriptomics
• Bioinformatics

Background:

• Long-read metatranscriptomics offers insights into microbial genetic diversity and gene expression.
• Challenges include lack of reference genomes and high sequencing error rates.
• Accurate full-length transcript characterization is crucial for understanding active microorganisms.

Purpose of the Study:

• To present Fungen, a novel reference-free tool for constructing accurate transcripts from long-read metatranscriptomic data.
• To overcome limitations of existing methods in handling sequence similarity and error rates.
• To enable high-resolution analysis of complex metatranscriptomes.

Main Methods:

• Fungen utilizes read clustering and error correction for transcript assembly.
• It operates without requiring high-quality reference genomes.
• The tool is designed for efficiency, reducing memory usage and improving speed.

Main Results:

• Fungen achieves superior accuracy in transcript determination compared to existing methods.
• Demonstrates a 22 to 56-fold speed improvement and reduced memory footprint.
• Successfully applied to marine and soil metatranscriptomic datasets for taxonomic and gene profiling.

Conclusions:

• Fungen provides a fast, scalable, and accurate solution for analyzing long-read metatranscriptomic data.
• Enables reliable gene clustering and sequence generation, even with high sequence similarity.
• Facilitates comprehensive understanding of eukaryotic microbial diversity and function.