R-Coffee

R-Coffee: Consistency-based ncRNA multiple sequence alignment with structural integration

R-Coffee performs multiple sequence alignment (MSA) of non-coding RNA (ncRNA) sequences by integrating predicted or known RNA secondary structures with primary sequence information using a consistency-based framework derived from T-Coffee.

Key Features:

• Secondary Structure Integration: Incorporates RNA secondary structure predicted by RNAplfold or provided experimentally to improve ncRNA alignment accuracy.
• Consistency-Based Alignment: Combines pairwise and multiple sequence alignment methods into a consensus library and refines alignments for consistency with structural constraints.
• Slow/Accurate Mode: Integrates Consan for highly accurate pairwise RNA alignment, optimized for small datasets (<5 sequences, <150–200 nucleotides).
• RM-Coffee Mode: Merges outputs from MUSCLE, MAFFT, and ProbConsRNA with R-Coffee to efficiently align larger datasets.
• Benchmark Validation: Evaluated on 388 BRAliBase reference datasets and 11 Cmfinder datasets, demonstrating high accuracy for short sequences and performance comparable to R-Coffee/Consan on larger datasets.

Scientific Applications:

• ncRNA Analysis: Supports studies of RNA function, evolution, and interaction networks by generating structure-aware alignments and refining outputs from existing sequence or structural aligners.

Methodology:

R-Coffee constructs a primary alignment library from multiple pairwise and MSA methods, integrates RNA secondary structure information predicted by RNAplfold or provided externally, and applies a consistency-based scoring scheme derived from T-Coffee to refine the final multiple sequence alignment under sequence and structural constraints.