StoatyDive: Evaluation and classification of peak profiles for sequencing data

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Peak-calling is crucial for analyzing cross-linking immunoprecipitation with high-throughput sequencing (CLIP-Seq) data.
• Existing methods lack tools to evaluate and classify predicted peaks based on their genomic profile shapes.
• This gap hinders downstream analyses like sequence motif and structure pattern prediction.

Purpose of the Study:

• To introduce StoatyDive, a novel tool for classifying and filtering CLIP-Seq peak shapes.
• To enhance the analysis of high-throughput sequencing data by evaluating peak profile diversity.
• To provide a quality control measure for CLIP-Seq experiments.

Main Methods:

• Development of StoatyDive for peak shape classification in CLIP-Seq data.
• Application of StoatyDive to classify peak shapes for the histone stem-loop-binding protein (SLBP).
• Comparative analysis of StoatyDive against existing tools for peak clustering.

Main Results:

• StoatyDive identified more distinct peak shape clusters in CLIP data compared to existing tools.
• Demonstrated StoatyDive's utility as a quality control and filtering tool for various CLIP-Seq datasets.
• Showcased that splicing-related proteins (RBM22, U2AF1) exhibit sharper peak shapes.

Conclusions:

• StoatyDive addresses the need for a peak shape clustering tool for CLIP-Seq data.
• The tool refines downstream analyses, including structure and sequence motif predictions.
• StoatyDive serves as an effective quality control measure for CLIP-Seq experiments.