[Cell molecular computer. IX. Coding principles in intracellular information processing].

Methods of coding the number and search for molecular program in a molecular computer are considered. The limited length of the nucleotide code is (see formula) where Pi -- probability of request of the given program, N -- total number of programs. Energetic expenditures for the synthesis of the code with the length l (in an ideal case without noise) E approximately 10 kT X l. Protein-nucleic recognition allows the work of the cell with almost the same expenditures on the account of Brown search in the presence of noise.