DART: a fast and accurate RNA-seq mapper with a partitioning strategy

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• RNA sequencing (RNA-Seq) enables high-resolution gene expression measurement but demands substantial computational resources.
• Accurate alignment of sequence fragments to a reference genome is a critical, computationally intensive step in RNA-Seq analysis.
• Existing de novo spliced RNA aligners face challenges, necessitating more efficient algorithms due to increasing sequencing data volumes.

Purpose of the Study:

• To introduce DART, a novel de novo mapping algorithm for RNA sequencing data.
• To present an efficient alignment strategy that overcomes the computational bottlenecks of traditional methods.
• To evaluate DART's performance against state-of-the-art RNA aligners.

Main Methods:

• Developed DART, a de novo RNA-seq mapping algorithm utilizing a partitioning strategy.
• The algorithm avoids the computationally expensive extension step common in other aligners.
• Performance was assessed using both synthetic and real next-generation sequencing (NGS) datasets.

Main Results:

• DART demonstrates high efficiency, requiring the least alignment time among compared methods.
• The algorithm achieves sensitivity and accuracy comparable to or exceeding state-of-the-art aligners.
• Experimental results validate DART's effectiveness on diverse RNA-Seq datasets.

Conclusions:

• DART offers a computationally efficient solution for de novo RNA-seq alignment.
• The partitioning strategy effectively reduces alignment time without compromising accuracy.
• DART represents a significant advancement for processing large-scale RNA-Seq data.