160 likes | 375 Views
Solving Double Digest Problem by Genetic Algorithm. Marek Kukačka & Zdeněk Pátek. Double Digest Problem (DDP). Input: Δ A – fragment lengths from the complete digest with enzyme A Δ B – fragment lengths from the complete digest with enzyme B
E N D
Solving Double Digest Problem by Genetic Algorithm Marek Kukačka & Zdeněk Pátek
Double Digest Problem (DDP) • Input: • ΔA–fragmentlengths from the complete digest with enzyme A • ΔB–fragment lengths from the complete digest with enzyme B • ΔAB–fragment lengths from the complete digest withbothAand B • Output: • A–location of the cuts in the restriction map for the enzyme A • B–location of the cuts in the restriction map for the enzyme B • Double Digest Problem is NP - complete
Mutation • Swap mutation • Insert mutation
Crossover • One-point crossover
Selection • Inverse Tournament Selection while (populationSize > N) do select individuals X1, X2 from population randomly if fitness(X1) > fitness(X2) remove X2 from population else remove X1 from population
Custom DDPs (1) • 1. Custom DDP (moderate) A: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 B: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 AB: 1, 2, 3, 4, 5, 4, 2, 6, 1, 6, 2, 4, 5, 4, 3, 2, 1 • 2. Custom DDP (easy) A: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 B: 1, 10, 1, 10, 1, 10, 1, 10, 1, 10 AB: 1, 10, 1, 1, 9, 1, 2, 8, 1, 3, 7, 1, 4, 6
Custom DDPs (2) • 3. Custom DDP (hard) A: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 B: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 AB: 1, 2, 3, 4, 2, 3, 6, 2, 5, 5, 3, 6, 3, 5, 5, 2, 6, 3, 2, 4, 3, 2, 1 • 4. Custom DDP (weird) A: (1, 3) 100 B: (2, 2) 100 AB: (1, 1, 2) 100