Fig. 1.
Fig. 1.

Some of the AlphaSim parameters that can be specified by the user, of which most can be changed during the course of a simulation. Asterisk denotes parameters that are immutable.

 


Fig. 2.
Fig. 2.

Principle of an AlphaSim simulation illustrated using a pedigree structured in four burn-in generations and one selection generation for two traits characterized by an additive genetic model. (1) Haplotype sequences and an internal pedigree are simulated. (2) Haplotypes are recombined and dropped into the base generation of the pedigree. At this step, single-nucleotide polymorphisms (SNPs) and quantitative trait nucleotides (QTNs) are selected. (3) An effect is assigned to each QTN, and, for each individual of the base generation of the pedigree, genetic values are calculated and phenotypes simulated. (4) Haplotypes of the base generation are recombined and dropped into the burn-in generations of the pedigree successively. Similar to the base generation, genetic values are calculated and phenotypes simulated for each individual of the burn-in generations. (5) A selection generation is simulated according to the selection method and strategy as defined by the user.

 


Fig. 3.
Fig. 3.

Output of AlphaSim by directory.

 


Fig. 4.
Fig. 4.

Simulation of the plant breeding pedigree in Example 1. The pedigree includes 12 burn-in generations (from founders to F10) and one selection generation (F11). Ten randomly selected founders are used to generate 10 double haploids (DHs). These DHs are crossed to simulate five unique F1 individuals. Selfing the F1 individuals results in 20 F2 individuals (i.e., four per one F1). The F2 genotypes are selfed through single-seed descent for eight generations to generate 20 F10 or recombinant inbred lines (RILs). Five RILs are then selected and selfed to create three F11 each.

 


Fig. 5.
Fig. 5.

Results of genotype × environment interaction simulated in Example 1. Five recombinant inbred lines (RILs) tested in three replicates in two contrasting environments with heritabilities of 0.8 and 0.2.

 


Fig. 6.
Fig. 6.

Simulation of the plant breeding pedigree in Example 2. The pedigree includes five burn-in generations (from founders to F3) and one selection generation. Ten randomly selected founders are used to generate 10 double haploids (DHs). These DHs are crossed to simulate five unique F1 genotypes. Selfing the F1 individuals result in 20 F2 individuals, four per F1. The F2 individuals are selfed through single-seed descent for one generation to simulate 20 F3, and the best performing F3 in each of the five biparental families is selected and selfed to create three F4 individuals.

 


Fig. 7.
Fig. 7.

Simulation of the development of F4 derived recombinant inbred lines (RILs) using single-seed descent combined with recurrent selection on F2 plants. The simulation included three scenarios differing from each other by the number of cycles of recurrent selection, which was 0, 2, or 4. The scenarios begin with a common pair of parents, which were crossed to generate the F1 plants. Selfing the F1 generated F2 plants. Recurrent selection consisted of selecting the two best performing F2 individuals based on their genomic estimated breeding values and crossing them to generate new F2 plants. The F4 RILs were then developed using single-seed descent.