MPGAfold

MPGAfold predicts RNA secondary structures using a massively parallel genetic algorithm to model folding pathways, capture intermediate and multiple conformational states, and evaluate co-transcriptional and complete folding dynamics relevant to RNA function.

Key Features:

• Massively parallel genetic algorithm (MPGA): Uses a genetic algorithm implemented in a massively parallel manner to explore RNA folding landscapes.
• Folding pathway simulation: Simulates folding pathways and dynamic behaviors including intermediate conformational states and transitions.
• Co-transcriptional and full-length analysis: Analyzes both co-transcriptional folding and complete RNA folding to identify structures arising during transcription and after completion.
• Stochastic ensemble and consensus analysis: Executes multiple stochastic runs to account for genetic algorithm variability and derive consensus insights.
• Population-size and family-level simulations: Performs multiple folding simulations across varying population sizes and across RNA sequences within a family.
• Data mining-based analysis: Applies data mining techniques to analyze outcomes from sequential, individual, and ensemble MPGAfold runs.
• Integration with StructureLab analysis workbench: Integrates results for downstream analysis within the StructureLab analysis workbench.

Scientific Applications:

• RNA secondary structure prediction beyond MFE: Identifies biologically relevant structures that may not correspond to minimum free energy conformations due to kinetic trapping or alternative functional states.
• Modeling co-transcriptional folding: Captures intermediate structures and folding trajectories that occur during transcription.
• Family-level conformer identification: Determines significant intermediate and final structures across multiple sequences within an RNA family.
• Studying folding dynamics and function: Investigates how folding kinetics and conformational states influence gene expression and molecular function.

Methodology:

Performs sequential and full-length MPGA runs using a massively parallel genetic algorithm, conducts multiple stochastic folding simulations across varying population sizes and RNA sequences, and applies data mining techniques to analyze simulation ensembles.