LncRNA-ID

LncRNA-ID evaluates the coding potential of transcripts, focusing on long non-coding RNAs (lncRNAs >200 nucleotides), by integrating sequence, ribosome profiling, and conservation features for transcript classification.

Key Features:

• Random forest algorithm: Implements a random forest classifier to predict transcript coding potential.
• Sequence characteristics of putative open reading frames (ORFs): Analyzes ORF sequence features to detect signals of protein-coding capacity.
• Translation scores from ribosomal coverage: Computes translation scores based on ribosome profiling (ribosomal coverage) to capture evidence of translation.
• Conservation against characterized protein families: Assesses evolutionary conservation by comparing sequences to characterized protein families to infer similarity to known proteins.

Scientific Applications:

• lncRNA annotation across species: Annotates lncRNAs across various species by distinguishing them from protein-coding transcripts.
• Transcript classification: Classifies transcripts by coding potential to separate lncRNAs from protein-coding genes.
• Support for functional studies: Provides classification evidence to inform studies of lncRNA functions and their regulatory roles in gene expression.

Methodology:

Applies a random forest machine-learning model that integrates ORF sequence-derived features, ribosome profiling-based translation scores, and conservation comparisons against characterized protein families.