Answers are grounded in this paper and cite their sources.
Synthetic DNA has emerged as a promising storage medium thanks to its exceptional information density and long-term stability. However, sequence-dependent biochemical properties introduced during the encoding process can significantly affect downstream molecular processing and data recovery. Among these properties, the thermodynamic stability of DNA sequences, commonly quantified by the Folding Gibbs free energy (ΔG), can influence amplification efficiency, sequencing coverage, and ultimately decoding performance. In this work, we experimentally investigate the impact of ΔG constraints applied during the encoding stage on DNA data recovery. We compare the reference JPEG DNA, the future image coding standard on DNA, with a ΔG-constrained variant using the same image encoded into synthetic DNA oligonucleotides. Both configurations were synthesized, amplified, sequenced, and decoded under identical experimental conditions. We analyze the relationship between predicted thermodynamic stability, sequencing coverage, and decoding performance to quantify the influence of sequence thermodynamics on the recovery of encoded information. Our results demonstrate that incorporating Δ G constraints during encoding substantially reduces the proportion of sequences with highly negative predicted Δ G values and improves sequencing coverage homogeneity, decreases the number of underrepresented oligonucleotides and dropouts, and consistently reduces the sequencing coverage required for successful decoding across three different synthesis providers, thereby reducing sequencing time. These findings provide experimental evidence that thermodynamic sequence optimization represents an effective design strategy for improving the robustness and efficiency of DNA data storage systems.
No takes yet. Share an insight, caveat, or question.
Mateos et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: