Physics-Informed Generative and Tensor-Based Framework for DNA Sequence Simulation and Genomic Structure Discovery

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2026 | Volume : 13 | Issue : 02 | Page :
    By

    Genius Walia,

  1. Assistant Professor, Department of Physics, FSH&L, Guru Kashi University, Talwandi Sabo, Bathinda, Punjab, India

Abstract

In this paper, we explore the intersection of artificial intelligence (AI) and mathematical physics to propose advanced methods for DNA sequence generation and analysis. Specifically, we investigate how physics-informed Generative Adversarial Networks (GANs) and tensor network representations can be harnessed to restructure DNA for applications in genetic science. The proposed methodology offers a unique integration of concepts of thermodynamic modeling with innovative GAN architecture in order to allow the creation of biologically relevant and scientifically valid genomic sequences. The inclusion of thermodynamic constraints in the learning process guarantees the maintenance of biological validity, structural stability, and functional significance of the created sequences. Besides, tensor decomposition methods are applied to uncover the latent structure and dimensions of genomic data in order to increase interpretability and computational performance during sequence analysis. The methodology allows one not only to improve the process of creating synthetic sequences, but also to analyze their entropy, energy landscape, and molecular interactions. As evidenced by simulation experiments, the suggested model demonstrates good results when capturing the complex structure of genomic data and maintaining biological dynamism during the synthesis process. At the same time, the application of artificial intelligence technologies in combination with concepts of statistical physics and thermodynamics provides many new opportunities for innovation in the field of synthetic biology, genomic engineering, drug discovery, and precision/personalized medicine.

Keywords: Artificial intelligence, Generative Adversarial Networks, Tensor Decomposition, DNA Sequence Generation, Deep Generative Models

[This article belongs to Research & Reviews: A Journal of Bioinformatics ]

How to cite this article:
Genius Walia. Physics-Informed Generative and Tensor-Based Framework for DNA Sequence Simulation and Genomic Structure Discovery. Research & Reviews: A Journal of Bioinformatics. 2026; 13(02):-.
How to cite this URL:
Genius Walia. Physics-Informed Generative and Tensor-Based Framework for DNA Sequence Simulation and Genomic Structure Discovery. Research & Reviews: A Journal of Bioinformatics. 2026; 13(02):-. Available from: https://journals.stmjournals.com/rrjobi/article=2026/view=244609


References

  1. Goodfellow IJ, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, et al. Generative adversarial networks. In: Proceedings of the 28th International Conference on Neural Information Processing Systems (NeurIPS 2014); 2014. p. 2672-2680. Available from: https://papers.nips.cc/paper/2014/hash/f033ed80deb0234979a61f95710dbe25-Abstract.html
  2. Killoran N, Lee LJ, Delong A, Duvenaud D, Frey BJ. Generating and designing DNA with deep generative models. arXiv [Preprint]. 2017. Available from: https://arxiv.org/abs/1712.06148
  3. Aghazadeh A, Nisonoff H, Ocal O, David H. Brookes, Huang Y, Koyluoglu OO, et al. Epistatic Net allows the sparse spectral regularization of deep neural networks for inferring fitness functions. Nature Communications. 2021;12(1):5225. doi:10.1038/s41467-021-25371-3.
  4. Gupta A, Zou J. Feedback GAN (FBGAN) for DNA: a novel feedback-loop architecture for optimizing protein functions. arXiv [Preprint]. 2019. Available from: https://arxiv.org/abs/1901.05790
  5. Stoudenmire EM, Schwab DJ. Supervised learning with tensor networks. In: Lee DD, Sugiyama M, Luxburg UV, Guyon I, Garnett R, editors. Advances in Neural Information Processing Systems 29 (NeurIPS 2016). Red Hook (NY): Curran Associates Inc.; 2016. p. 4799-4807. Available from: https://papers.nips.cc/paper/2016/hash/5314b9674c86e3f9d1ba25ef9bb32895-Abstract.html
  6. Zhang Y, Li X, Liu T. Tensor decomposition reveals deep phenotype associations in DNA methylation data. Bioinformatics. 2020;36(5):1470-1476. doi:10.1093/bioinformatics/btz703. Available from: https://doi.org/10.1093/bioinformatics/btz703
  7. Kingma DP, Ba J. Adam: a method for stochastic optimization. arXiv [Preprint]. 2014. Available from: https://arxiv.org/abs/1412.6980
  8. Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv [Preprint]. 2015. Available from: https://arxiv.org/abs/1511.06434
  9. Ghosh A, Zhang W, Song L. From generative models to physical systems: physics-informed machine learning. Nat Comput Sci. 2021;1(1):1-7. doi:10.1038/s43588-020-00001-4. Available from: https://doi.org/10.1038/s43588-020-00001-4
  10. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596(7873):583-589. doi:10.1038/s41586-021-03819-2. Available from: https://doi.org/10.1038/s41586-021-03819-2
  11. Acar E, Dunlavy DM, Kolda TG, Mørup M. Scalable tensor factorizations for incomplete data. Chemom Intell Lab Syst. 2011;106(1):41-56. doi:10.1016/j.chemolab.2010.08.004. Available from: https://doi.org/10.1016/j.chemolab.2010.08.004
Regular Issue Subscription Review Article
Volume 13
Issue 02
Received 29/04/2026
Accepted 18/05/2026
Published 21/05/2026
Publication Time 22 Days
11

Login


My IP

PlumX Metrics