Minimal protocell concepts have high intellectual and practical value. Following chemoton theory, developed by Tibor Gánti in 1971, we suggest that a minimal protocell satisfying all life criteria should consist of three subsystems: a metabolic network producing materials for the production of all three subsystems at the expense of the difference between food and waste products, a genetic subsystem based on template polymerization and a boundary subsystem consisting of a bilayer vesicle. All three subsystems are autocatalytic and the system as a whole (called chemoton) is also autocatalytic. The chemoton can undergo spatial reproduction in the right parameter domain. Various infrabiological systems can be constructed from any two subsystems; we review the experimental attempts aimed at such a goal. As the complexity of the realized systems increases, the problem of unwanted side reactions becomes more and more dramatic in non-enzymatic systems. No satisfactory solution to the problem of metabolite channeling is known.