A scalable and accurate targeted gene assembly tool (SAT-Assembler) for next-generation sequencing data

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• De novo gene assembly from next-generation sequencing (NGS) data is crucial for functional analysis, especially for non-model organisms and metagenomics.
• Existing assembly tools struggle with challenges like heterogeneous coverage, gene/isoform similarity, and large data sizes, leading to fragmented or chimeric contigs and high memory usage.
• Reference-free assembly is essential when high-quality reference genomes are unavailable.

Purpose of the Study:

• To introduce SAT-Assembler, a novel targeted gene assembly program designed to efficiently recover specific gene families of biological interest.
• To address the limitations of current de novo assembly methods, including fragmentation, chimerism, and high computational resource demands.
• To provide a flexible tool applicable to both RNA-Seq and metagenomic datasets.

Main Methods:

• Development of SAT-Assembler utilizing family-specific homology search for targeted gene recovery.
• Construction of homology-guided overlap graphs to improve assembly accuracy.
• Application of careful graph traversal algorithms to minimize errors and enhance contig quality.
• Evaluation on Arabidopsis RNA-Seq and two metagenomic datasets.

Main Results:

• SAT-Assembler demonstrated reduced memory footprint compared to existing tools.
• Achieved comparable or superior gene coverage in targeted gene family assembly.
• Exhibited a lower rate of chimeric contigs, indicating improved assembly accuracy.
• Showcased compatibility with parallel computing due to its design.

Conclusions:

• SAT-Assembler offers an effective solution for targeted gene family assembly, overcoming key challenges in de novo sequencing data analysis.
• The tool provides a more memory-efficient and accurate alternative for researchers working with RNA-Seq and metagenomic data.
• Its family-specific approach and parallel computing compatibility enhance its utility in biological research.