Numerous diseases in humans have been associated with mutations of the mitochondrial genome (mtDNA). This genome encodes 13 protein subunits of complexes involved in oxidative phosphorylation (OXPHOS), a process that provides aerobic eukaryotes with the energy-rich adenosine triphosphate molecule (ATP). Mutations of the mtDNA may therefore have dramatic consequences especially in tissues and organs with high energy demand. Evaluating the pathogenicity of these mutations may be difficult because they often affect only a fraction of the numerous copies of the mitochondrial genome (up to several thousands in a single cell), which is referred to as heteroplasmy. Furthermore, due to its exposure to reactive oxygen species (ROS) produced in mitochondria, the mtDNA is prone to mutations, and some may be simply neutral polymorphisms with no detrimental consequences on human health. Another difficulty is the absence of methods for genetically transforming human mitochondria. Face to these complexities, the yeast Saccharomyces cerevisiae provides a convenient model for investigating the consequences of human mtDNA mutations in a defined genetic background. Owing to its good fermentation capacity, it can survive the loss of OXPHOS, its mitochondrial genome can be manipulated, and genetic heterogeneity in its mitochondria is unstable. Taking advantage of these unique attributes, we herein describe a method we have developed for creating yeast models of mitochondrial ATP6 gene mutations detected in patients, to determine how they impact OXPHOS. Additionally, we describe how these models can be used to discover molecules with therapeutic potential.