In the first part, I show how digital organisms adapt their genomes to encode cooperation-related genes in a more constrained way (evolvability suppression), especially using operons and overlaps also involving essential genes. In the second part, we experimentally test this view of gene overlaps as an evolutionary constraint, using both algorithmic and synthetic biology tools that we have developed. In the third part, I use agent-based simulations to show how a form of division of labour can be interpreted as a cooperative system in the light of modern evolutionary theory. In the final part, I show that the patterns of dispersal of cooperative alleles due to hitchhiking phenomena play an important role in the evolution of cooperation. The last result holds even though the hitchhiking mechanisms also applies to non-cooperative alleles, thanks to the relatedness (at cooperation-related loci) created by the local invasion of beneficial mutations (at loci not related to cooperation). The beneficial mutations form a complex and interesting equilibrium with mutational robustness, which I investigate using in silico evolution. On the whole, these results call for a more careful consideration of the second-order selection pressures in the study of social evolution, and show the necessity for more realistic models allowing to integrate such evolutionary forces. My thesis research specifically highlights the importance of the mutational landscape in the study of microbial populations and shows the increasing potential of synthetic biology as a tool to study such landscape and microbial evolution in general.