microDoR

microDoR predicts mechanisms of microRNA (miRNA)-mediated gene silencing in human cells by discriminating between mRNA degradation (cleavage) and translational repression.

Key Features:

• Online SVM-based machine learning: Utilizes an online Support Vector Machines (SVM) algorithm to classify miRNA regulatory outcomes.
• miRNA-mRNA duplex structure: Considers the duplex conformation of the miRNA-mRNA complex as a predictive feature.
• Binding site count: Incorporates the number of miRNA binding sites on the target mRNA.
• Structural accessibility: Evaluates structural accessibility of target site regions on the mRNA.
• Outcome prediction: Predicts whether a target mRNA will be cleaved (degraded) or translationally inhibited by miRNAs.

Scientific Applications:

• Mechanism assignment: Assigns likely silencing mechanisms (cleavage versus translational repression) to specific miRNA-mRNA pairs in human cells.
• Regulatory determinant analysis: Analyzes how duplex structure, binding site number, and site accessibility influence miRNA regulatory mode.

Methodology:

Uses an online Support Vector Machine (SVM) model that analyzes miRNA-mRNA duplex structure, number of binding sites on the mRNA, and structural accessibility of target site regions to predict cleavage versus translational inhibition.