Branchpointer

Branchpointer predicts branchpoint probability within intronic regions to annotate branchpoint elements and assess the impact of single nucleotide polymorphisms (SNPs) on splicing.

Key Features:

• Machine Learning Algorithm: Employs a machine-learning algorithm to identify branchpoint elements from gene annotations and genomic sequences.
• Performance: Achieves sensitivity of 61.8% and specificity of 97.8%, and annotates branchpoints in approximately 85% of human gene introns.
• Mutation Impact Evaluation: Evaluates the impact of single nucleotide polymorphisms (SNPs) on branchpoint architecture to interpret potential splicing consequences.
• Clinical Variant Analysis: Identifies known deleterious branchpoint mutations documented in clinical variant databases and predicts thousands of additional clinical and common variants that may affect branchpoint function.
• Genome-wide Annotation: Provides comprehensive genome-wide annotation of branchpoints as a reference for splicing studies and noncoding variant interpretation.
• Input Data: Operates using gene annotations and genomic sequences as input.

Scientific Applications:

• Splicing mechanism research: Annotation of branchpoints to support studies of spliceosome recognition and splicing mechanisms in eukaryotic genes.
• Variant interpretation and clinical genetics: Assessment of SNP effects on branchpoints to aid interpretation of noncoding variants and clinical variant evaluation.

Methodology:

Applies a machine-learning algorithm to predict branchpoint probability from gene annotations and genomic sequences and assesses SNP-induced changes in predicted branchpoint architecture.