GERP++

GERP++ identifies functionally constrained elements within genomes by analyzing multiple sequence alignments to detect regions under selective constraint.

Key Features:

• Maximum likelihood evolutionary rate estimation: Uses maximum likelihood to estimate evolutionary rates for position-specific scoring.
• Dynamic programming breakpoint evaluation: Employs dynamic programming to evaluate a broad range of candidate breakpoints instead of heuristic extension.
• Position-specific constraint scores: Calculates constraint scores at individual nucleotide positions within multiple sequence alignments.
• Constrained element definition: Defines constrained elements as contiguous segments of highly scored positions.
• Statistical ranking of elements: Ranks candidate constrained elements based on statistical significance.
• Nucleotide- and element-level annotation: Provides both nucleotide-level and element-level constraint scores within deep multiple sequence alignments.
• Comparative genomics focus: Leverages comparative genomics to detect selective constraint indicative of functional importance.
• Genome-scale results: Identified over 1.3 million constrained elements covering more than 7% of the human genome.

Scientific Applications:

• Detection of constrained genomic regions: Identifies genomic regions under purifying selection that are likely to be functionally important.
• Improved correspondence with known functional sequences: Enhances mapping between predicted constrained elements and known functional sequences by annotating longer elements.
• Comparative analyses across species: Enables comparative genomics analyses using deep multiple sequence alignments to assess constraint.
• Annotation of extended functional elements: Facilitates annotation of longer constrained elements to capture extended functional sequences.

Methodology:

Analyzes multiple sequence alignments; applies maximum likelihood evolutionary rate estimation for position-specific scoring; uses dynamic programming to evaluate and statistically rank candidate breakpoints; calculates nucleotide-level constraint scores and defines constrained elements as contiguous high-scoring segments.