ADAM

ADAM is a modular, stochastic-simulation program for evaluating animal breeding schemes under a wide range of genetic architectures and management scenarios. It simulates a user-specified base population and descendant generations across discrete time steps, using alternative genetic models for single or multiple traits: (i) an infinitesimal (polygenic) model, (ii) a finite-locus model representing genes and/or markers, or (iii) a mixed-inheritance model combining both. Users parameterize population structure (e.g., reproductive ages, trait expression timing, germplasm availability, herd structure) and define selection and mating decisions at each time step, including multi-stage selection across strata (sex, age, herd, families) and hierarchical or factorial mating within/across herds.

ADAM supports both conventional and molecularly informed strategies, including marker- and gene-assisted selection, selective genotyping, MOET, and sexed semen. Selection can be random, truncation-based, or optimum-contribution, using criteria such as phenotype, BLUP (phenotype+pedigree), MA-BLUP (marker-informed), and GA-BLUP (gene-informed). Mating options include random mating, minimum coancestry, and minimization of covariance of ancestral contributions.

Outputs include true and estimated breeding values per time step, trait-wise and overall genetic gain, generation interval, prediction accuracy, and inbreeding metrics derived from pedigree data and optionally DNA data (including lengths of fixed genomic regions). Implemented in FORTRAN 95, ADAM integrates external components for key procedures (random number generation, breeding value estimation, optimum-contribution selection, and identity-by-descent calculations), and is designed to be extensible to new breeding technologies and strategies.