EMSAR: estimation of transcript abundance from RNA-seq data by mappability-based segmentation and reclustering.

Area of Science:

• Bioinformatics
• Genomics
• Computational Biology

Background:

• RNA-sequencing (RNA-seq) is crucial for genome-wide expression profiling.
• Ambiguity in read mapping arises from sequence similarity among genes and isoforms.
• Current RNA-seq quantification methods often face trade-offs between accuracy and computational efficiency.

Purpose of the Study:

• To introduce EMSAR (Estimation by Mappability-based Segmentation And Reclustering), a novel computational method for accurate transcript abundance quantification from RNA-seq data.
• To address the limitations of existing RNA-seq analysis tools regarding accuracy and computational cost.

Main Methods:

• EMSAR groups reads based on their mapped transcript sets.
• It employs a joint Poisson model for maximum likelihood estimation within optimal transcript segments.
• Utilizes efficient transcriptome indexing via modified suffix arrays for optimized performance.

Main Results:

• EMSAR effectively quantifies transcript abundance using a large proportion of mapped reads, including ambiguously mapped ones.
• The method achieves high accuracy comparable to leading RNA-seq quantification tools.
• Demonstrates minimized CPU time and memory usage.

Conclusions:

• EMSAR provides a highly accurate and computationally efficient solution for transcript quantification in RNA-seq data analysis.
• This method enhances genome-wide expression profiling capabilities.
• The EMSAR tool is publicly available for research use.