Over the last 50 years, we have focused most of our research on the quantification of human variability in response to acute and chronic exposure to aerobic exercise as well as alterations in energy balance conditions, and on the potential causes of the observed human heterogeneity. In the investigation of an exercise biology trait, it is essential to distinguish between the sedentary state and the changes in the same trait in response to an exercise program or exercise training. Using maximal oxygen uptake (VO2max) adjusted for body mass and composition as an example, we have found that even among sedentary adults of the same sex and ethnic background, and within a limited age range, there is at least a two-fold range between very low and very high VO2max individuals. This variability is not random as about 50% of the variance is accounted for by genetic differences as shown in twin and family studies. Cardiac output, blood volume and hemoglobin content, skeletal muscle oxidative capacity and transcriptomic profile, genomic variants, plasma secreted proteins and metabolites have been associated with VO2max in the sedentary state but much remains to be learned about the nature of individual differences in the sedentary state. In a series of experiments, we studied the VO2max response (trainability) to a number of exercise regimens in biologically unrelated adults and in pairs of identical twins and nuclear families. First, we repeatedly found that there is no correlation between VO2max in the sedentary state and its trainability. Second, there is at least a 3-fold range between the low and high VO2max gainers in response to a standardized endurance-training program. This is true in young and middle-aged adults, men or women, and in people of African or European ancestry. Third, the heritability of VO2max training response reaches about 50% of the remaining variance after adjustment for relevant concomitants. Fourth, genome-wide screens, vastus lateralis muscle transcripts abundance profiling, extensive proteomics and metabolomics exploration, and comprehensive bioinformatics pipelines have allowed to identify multiple molecular makers and pathways associated with human variation in VO2max trainability. Fifth, DNA variants and protein levels significantly associated with VO2max in the sedentary state and its response to exercise training are almost entirely different supporting the notion that these two traits are truly independent from one another. Sixth, the molecular and pathway profiles of VO2max in the sedentary state are well aligned with its recognized physiological determinants (cardiac output, blood volume and hematopoiesis, skeletal muscle oxidative capacity, etc.). In contrast, they differ markedly for VO2max trainability as they are dominated by developmental and embryonic, regulation of gene expression, angiogenesis, skeletal and cardiac muscle growth, extracellular matrix, regulation of apoptosis and autophagy, immunity, calcium signaling and energy metabolism pathways. Even though the notion of human variability of exercise-related traits is now almost universally accepted, we have a long way to go before a coherent description of the molecular and physiological basis for this variability is achieved.