Fig 1.
A digital system where phenotypic changes can be incorporated into genotype.
In this system, sound is stored directly in DNA. A table that relates note pitch and duration combinations to 4-letter codons (see text for details) provides a means of converting between information stored as synthetic DNA (genotype) and sound (phenotype). It is possible to cycle between these two states, as follows. In Step 1, musical notation is converted to DNA sequence (reverse translation) using a conversion table that links musical notes and durations to DNA codons. The resulting DNA sequence file is used to guide synthesis of physical DNA (Step 2). The information within the synthetic DNA can be read by Sanger sequencing, generating a digital sequence file (Step 3). The resulting DNA sequence file is then ‘translated’ to musical notation using the conversion table, yielding a musical score in both MIDI and WAV formats (Step 4). Steps 2 & 3 can be skipped, allowing the DNA sequence file (FASTA) from Step 1 to be translated into musical notation (Step 2A, which is equivalent to Step 4). In Steps 5–7, audio is played through loudspeakers (Step 5), captured through recording (Step 6) and converted back into a musical score (midi file, Step 7). Created in BioRender. Poole, A. (2025) https://BioRender.com/g7q2rjf.
Table 1.
Mutational regimes.
Fig 2.
Comparison of mean Hamming distance between the ancestor and descendent generations under a range of mutational regimes.
Each regime (Table 1) was iterated ten times across 20 generations, and Hamming distances were calculated at each generation for each replicate by comparison to the ancestor sequence. Regime 1: ‘zero’ mutation control (1_zero; black line); Regime 2: synonymous DNA-level mutation (2_synDNA; blue line); Regime 3: nonsynonymous DNA-level mutation (3_nonsynDNA; purple line); Regime 4: random DNA-level mutation (4_ranDNA; green line); Regime 5: nonsynonymous music-level mutation (5_nonsynMusic; orange line); Regime 6: nonsynonymous DNA-level mutation plus nonsynonymous music-level mutation (6_nonsynDNA+Music; red line). The x-axis shows generation number, y-axis shows mean Hamming distances for each regime calculated at each generation from comparisons between ancestor and the ten iterations.
Fig 3.
Effect of selection regime on evolution of DNA sequence and musical score.
A. Bitscore comparison of DNA sequences selected by majority wins human poll (blue) versus random choice to mimic neutral evolution (orange) for four rounds of evolution (R1-R4). B. Bitscore comparisons between ancestor musical score (A) and human poll (blue) or random choice (orange) for R1-R4. The first data point represents the music-score comparison of ancestor against itself to establish the maximum similarity score (310).