Shark: fishing relevant reads in an RNA-Seq sample

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• High-throughput RNA-Seq generates massive datasets, often containing irrelevant reads.
• Irrelevant reads can significantly degrade the performance of data analysis tools.
• Focusing on specific genes necessitates efficient filtering of non-target reads.

Purpose of the Study:

• To introduce a novel computational problem: gene assignment.
• To develop an efficient, alignment-free method for gene assignment in RNA-Seq data.
• To assess the impact of gene assignment on the performance of RNA-Seq analysis pipelines, particularly for differential splicing.

Main Methods:

• Proposed an alignment-free approach to solve the gene assignment problem.
• Developed a tool named Shark to implement the gene assignment method.
• Assessed Shark's effectiveness in speeding up differential splicing analysis pipelines.

Main Results:

• Shark successfully extracts relevant reads from RNA-Seq samples for a given gene panel.
• The tool handles samples with novel alternative splicing events effectively.
• Shark significantly improves the performance of RNA-Seq analysis tools.

Conclusions:

• Gene assignment is a crucial step for optimizing RNA-Seq data analysis.
• Shark provides an efficient solution for gene assignment, enhancing analysis speed.
• The implemented tool, Shark, improves performance without compromising the accuracy of RNA-Seq study results.