CarthaGene 1.3 beta

CarthaGene 1.3 beta constructs multipoint maximum likelihood consensus maps for genetic and radiated hybrid mapping to order markers and integrate pooled data from multiple crosses.

Key Features:

• Multipoint maximum likelihood consensus maps: Builds accurate multipoint maximum likelihood consensus maps for genetic and radiated hybrid data.
• Automatic marker ordering: Automatically orders markers to generate consensus maps.
• Integration of pooled cross data: Integrates pooled data from multiple crosses.
• Hybrid EM and local search algorithm: Combines the Expectation-Maximization (EM) statistical optimization technique with local search strategies derived from artificial intelligence and operations research.
• Estimation of recombination fractions: Estimates maximum likelihood recombination fractions efficiently.
• TSP-inspired local search for marker ordering: Uses a structured local search process inspired by the traveling salesman problem to optimize marker order.
• Support for diverse pedigrees and map joining: Joins genetic and radiated hybrid maps across backcrosses, recombinant inbred (RI) lines, F2 intercrosses, phase-known outbreds, and both haploid and diploid radiation hybrids.

Scientific Applications:

• Marker ordering resolution: Addresses marker ordering issues in linkage and radiation hybrid mapping.
• Map integration across pedigrees: Joins genetic and radiated hybrid maps across multiple crosses and pedigrees for consensus map construction.
• QTL and gene localization: Facilitates precise localization of genes or quantitative trait loci (QTL) through robust maximum likelihood map construction.
• Empirical validation: Applied to the wasp Trichogramma brassicae and random pooled datasets, demonstrating superior maximum likelihood outcomes compared to existing mapping software.

Methodology:

Uses a hybrid algorithm combining Expectation-Maximization (EM) with local search strategies from artificial intelligence and operations research, employing a structured local search inspired by the traveling salesman problem to estimate maximum likelihood recombination fractions, optimize marker order, and integrate pooled cross data.