IRFinder-S

IRFinder-S quantifies and detects intron retention events across multiple samples using convolutional neural networks to distinguish genuine intron retention from other genomic features and artifacts.

Key Features:

• Quantification and detection: Accurately quantifies and detects intron retention events across multiple samples.
• Convolutional neural networks (CNNs): Integrates CNNs to improve accuracy by distinguishing genuine intron retention from other genomic features or artifacts.
• Collaborative data sharing: Facilitates exchange of intron retention results among research laboratories for cross-study comparison and validation.
• Dynamic sample database: Provides a dynamic database to explore and contrast available samples and compare intron retention levels between conditions or experimental groups.
• Differential intron retention detection: Implements a validated method for detecting differential levels of intron retention between conditions.

Scientific Applications:

• Comparative analysis: Compare intron retention levels between different conditions or experimental groups to identify condition-specific retention.
• Differential retention studies: Detect differential intron retention associated with treatments or biological perturbations.
• Regulatory mechanism investigation: Investigate regulatory mechanisms and functional implications of intron retention in gene expression.
• Cross-study validation: Enable validation of intron retention findings across laboratories and studies.

Methodology:

Uses convolutional neural networks (CNNs) to distinguish genuine intron retention from other genomic features or artifacts, quantifies intron retention across multiple samples, maintains a dynamic database for sample comparison, and applies a validated method to detect differential intron retention.