CirRNAPL

CirRNAPL identifies circular RNAs (circRNAs) and analyzes their expression by classifying circRNA sequences using an extreme learning machine optimized for accurate circRNA identification.

Key Features:

• Feature Extraction: Extracts compositional and structural features from nucleic acid sequences of circRNAs for use in classification.
• Optimized Machine Learning Algorithm: Uses an extreme learning machine (ELM) whose parameters are optimized via particle swarm optimization (PSO) to improve predictive accuracy and efficiency.
• Comparative Performance: Benchmarked against existing methods including BLAST across three datasets, reporting accuracies of 0.815, 0.802, and 0.782 respectively.
• Expression Analysis: Incorporates sequence alignment on an independent detection set to evaluate expression levels and reports a positive correlation between sequence abundance and expression.

Scientific Applications:

• CircRNA identification: Enables accurate detection of circRNAs from nucleotide sequence data for genomics and transcriptomics studies.
• Disease mechanism studies: Supports investigation of circRNAs' roles in disease development by providing reliable identification and expression information.
• Drug development and therapeutics: Facilitates exploration of circRNAs as potential drug targets and therapeutic biomarkers through identification and expression analysis.

Methodology:

Extracts compositional and structural sequence features; classifies sequences using an extreme learning machine optimized by particle swarm optimization (PSO); performs sequence alignment on an independent detection set; benchmarks against BLAST with reported accuracies of 0.815, 0.802, and 0.782 on three datasets.