Merlin--rapid analysis of dense genetic maps using sparse gene flow trees

Area of Science:

• Genetics
• Computational Biology
• Bioinformatics

Background:

• High-density single-nucleotide polymorphism (SNP) maps generate large datasets challenging current computational tools for disease gene discovery.
• Efficient analysis of genetic maps in pedigree data is crucial for identifying disease-associated genes.

Purpose of the Study:

• To introduce a novel, efficient computational method for analyzing dense genetic maps in pedigree data.
• To provide fast solutions for allele-sharing analyses and haplotyping in complex pedigrees.

Main Methods:

• Development of sparse binary trees to represent gene flow patterns in pedigrees.
• Derivation of algorithms for pedigree traversal and efficient likelihood calculations.
• Implementation of an approximate multipoint calculation for enhanced speed in dense maps.
• Creation of the Multipoint Engine for Rapid Likelihood Inference (Merlin) software.

Main Results:

• Sparse binary trees offer a parsimonious representation of gene flow.
• Exact likelihood calculations are efficient for single and multiple linked markers.
• Approximate multipoint calculations provide accurate solutions for dense maps with thousands of markers.
• Merlin demonstrates superior performance in speed and marker handling compared to existing packages.

Conclusions:

• The developed method and Merlin software significantly enhance the efficiency of genetic map analysis in large pedigrees.
• This approach facilitates faster and more accurate disease gene mapping and haplotyping.
• Merlin provides a powerful tool for genotype error detection and affected pair linkage analyses.