Complex population demography can have subtle yet significant impacts on the genetic variation of populations. Furthermore, complex demography can subtly affect natural selection and therefore shapes the distribution of deleterious genetic variation. In my dissertation, I utilize a variety of computational tools to model the impact of deleterious variation in complex populations. In the first chapter, I investigated why among all human populations East Asians have the most Neanderthal ancestry. I found that multiple interbreeding events between Neanderthals and East Asians are required to explain the data, revising current models of human history. In the second chapter, I developed new computational tools for estimating the distribution of fitness effects using large datasets of genetic variants and estimated the amount of selection on amino acid changing mutations in humans. Here I found fewer strongly deleterious mutations compared to previous smaller studies, suggesting that neutral forces may play a greater role in human evolution than previously appreciated. In Chapter 3, I investigated the dynamics of deleterious genetic variation in hybrid populations using simulations and found that differences in standing deleterious variation between parent populations can significantly impact the evolution of hybrids. It is therefore essential that null models of hybrid evolution consider the effects of deleterious variation before invoking processes such as hybrid incompatibility or adaptive introgression to explain unusual patterns of genetic variation.