Genes code for 20 different amino acids that occur in proteins. There's genetically combinatorially room for more (64, right?) and hundreds of different (today abiotic) amino acids are chemically possible. Is there any evidence or theory about how the number and relative frequency of different proteinogenic amino acids in the biosphere has changed throughout the evolution? For example, have they increased slowly or suddenly or has there been these 20 of them for as long as can be determined?