Fig. 9
The translation process in Aevol. a Aevol’s genetic code. The code contains 23=8 non-redundant codons, two of which encoding for START and STOP (000 and 001 respectively). The six remaining codons encode for six “Amino-Acids” (AA) that are pairwise associated to three classes respectively coding for M (AA M0 and M1), W (AA W0 and W1) and H (AA H0 and H1). b Translation of the gene into the protein’s primary structure. The transcribed RNA sequence is parsed for Ribosome Binding Sites (RBS) followed by a START codon. The translation then started on the reading frame of the START codon until a STOP codon is found (note that the length of the gene is not predefined). The protein’s primary structure is a sequence of AA whose length depends on the gene’s length. c Computation of the protein’s parameters. The AA sequence is parsed into three substructures containing respectively all AA of class “M”, “W” and “H”. These substructures are then translated into binary sequences that are converted into an integer value via Grey code. This value is turned into a fraction that determines the actual value extracted from the range of possible values of that property. d Graphical representation of the protein function. The protein is represented as a triangle defined by three properties: Main phenotypic trait (M), level of contribution to the trait (H) and level of pleiotropy (W). For sake of clarity, the phenotypic target is also represented on the figure (see Fig. 8). Note that the H value may be scaled by the activity of the gene’s promoter