CircCode

CircCode identifies the coding potential of circular RNAs (circRNAs) by analyzing ribosome profiling data using Python 3.

Key Features:

• Automated Pipeline: Implements sequence linking, quality control, filtering, and alignment as computational steps for circRNA translation analysis.
• Machine Learning Integration: Uses Random Forest and J48 classification algorithms to predict circRNA coding ability.
• Implementation: Developed in Python 3.
• Performance: Reported to achieve low false discovery rate and high sensitivity for identifying translated circRNAs.

Scientific Applications:

• Cross-species Identification: Identified 3,610 translated circRNAs in humans and 1,569 in Arabidopsis thaliana.
• Functional Investigation: Enables exploration of circRNA regulatory roles beyond serving as microRNA (miRNA) decoys.

Methodology:

Computational steps explicitly include sequence linking, quality control, filtering, alignment, analysis of ribosome profiling data, processing ribosome profile databases from NCBI, and classification using Random Forest and J48 algorithms.