Programs for simulation of the evolution of recombination

Program simulating effect of recombination for Felsenstein 1974 paper

This is a program written by me for the 1974 paper on the evolutionary advantage of recombination. It simulates the fixation of advantageous alleles in the absence of recombination and compares that to the expected rate of fixation at loci that are not in linkage disequilibrium (computed from Kimura's 1962 formula). It is written in FORTRAN IV for the University of Washington's CDC 6400 mainframe computer.

Click here to display or download recomb1974.for

Programs simulating selection for recombination for the Felsenstein and Yokoyama 1976 paper

These two programs were written for the 1976 paper by Felsenstein and Yokoyama in Genetics. I designed the programs but I think that Shozo Yokoyama wrote them, and he did the runs that were reported in the paper.

They are written in FORTRAN IV for the University of Washington's CDC 6400 mainframe. A warning about compiling the FORTRAN IV programs: They use the operator .AND. to do bitwise AND operations. This may work in FORTRAN 77, as there is backwards compatibility in it. In Gnu FORTRAN it will work if you use the -fdec flag with the compiler. Otherwise it may be necessary to change expressions like M.AND.N to IAND(M, N) or to AND(M,N).

One case that was run was where the locus that controls recombination had the allele for recombination dominant, the other had it recessive. Generally they were run with an equal initial frequency of the recombination allele (50:50) and then we could detect the effectiveness of natural selection for recombination by finding that the recombination allele fixed significantly more than 50% of the time.

I believe that recomb1976a.for simulates the dominant case, and recomb1976b.for the recessive case. In the latter case the two subpopulations that have the two different alleles at the recombination locus have no gene flow between them, so that Fisher and Muller's argument for the advantage of recombination predicts that selection for recombination will occur, as the subpopulation with more recombination will accumulate more favorable mutants. It is less obvious that this will be true in the dominant case, but it proved to be true in the simulations.

Click here to display or download recomb1976a.for, and

Click here to display or download recomb1976b.for.