COME

COME predicts the coding potential and provides characterization of novel long non-coding RNAs (lncRNAs) from RNA sequencing (RNA-seq) data by integrating multiple sequence-derived and experiment-based features.

Key Features:

• Decompose–compose integration: Implements a decompose–compose method to integrate diverse sequence-derived and experiment-based features for coding-potential calculation.
• Comprehensive annotation: Provides per-transcript annotation and characterization supported by multiple lines of evidence.
• Conserved local secondary structures: Identifies conserved local RNA secondary structures within lncRNAs, with reported enrichment in lncRNAdb and increased interactions with RNA binding proteins.
• Improved prediction consistency: Enhances accuracy and robustness of coding-potential predictions and consistency of results relative to other coding potential calculators.

Scientific Applications:

• lncRNA discovery: Identification of novel lncRNA transcripts from RNA-seq datasets.
• Functional characterization: Characterization of lncRNAs to support studies of regulatory mechanisms and interactions with RNA binding proteins.
• Transcriptome annotation: Generation of detailed annotations to support transcript-level analyses and integration with databases such as lncRNAdb.

Methodology:

COME applies a decompose–compose method to integrate sequence-derived and experiment-based features, identifies conserved local RNA secondary structures, and supports predictions with multiple lines of evidence.