GENERATIVE AI FOR DE NOVO DRUG DESIGN: NEW CHALLENGES IN MOLECULE
Keywords:
generative models, biological compounds, small molecules, datasets, bench marks, model architecture, quantum computing, QSARAbstract
Artificial intelligence-based methods can significantly improve the traditional expen sive drug development process, given the fact that various generative models are already widely used in chemistry. Generative models for de novo drug design are focused on cre ating new biological compounds completely from scratch, which represents a promising direction in the future. The rapid development in this field, combined with the inherent complexity of the drug design process, creates difficult conditions for researchers. Within the framework of the topic of creating small molecules, we define many subtasks and applications, highlighting important datasets, benchmarks, model architecture and com paring the performance of the best models. The review presents key advances in this f ield, including the advent of quantum computing, which promises to further accelerate the application of deep QSAR to support computer-aided drug design in the field of molecules.
References
Tropsha A. et al. Integrating QSAR modelling and deep learning in drug discovery: the emergence of deep QSAR // Nature Reviews Drug Discovery.– 2024.– Vol. 23.– P. 141–155.– doi: http://dx.doi.org/10.1038/s41573-023-00832-0.
Tang X. et al. A Survey of Generative AI for de novo Drug Design: New Frontiers in Molecule and Protein Generation //arXiv.– 2024.– https://arxiv.org/html/2402.08703v1.
Wang M. et al. Deep learning approaches for de novo drug design: An overview // Current Opinion in Structural Biology.– 2022.– Vol. 72.– P. 135-144.– doi: http://dx.doi.org/ 10.1016/j.sbi.2021.10.001.
Liu X. et al. Computational approaches for de novo drug design: past, present, and future // Methods Mol Biol..– 2021.– Vol. 2190.– P. 139-165.– doi: http://dx.doi.org/10.1007/ 978-1-0716-0826-5_6.
DiMasi J.A., Grabowski H.G., Hansen R.W. Innovation in the pharmaceutical industry: new estimates of r&d costs // Journal of health economics.– 2016.– Vol. 47.– P. 20-33.
Jayatunga M. et al. AI in small-molecule drug discovery: a coming wave? // Nature Reviews Drug Discovery.– 2022.– Vol. 21.– P. 175-176.
Zhang M. et al. A survey on graph diffusion models: Generative ai in science for molecule, protein and material // arXiv.– 2023.– https://arxiv.org/abs/2304.01565.
Guo Z. et al. Diffusion models in bioinformatics: A new wave of deep learning revolution in action // arXiv.– 2023.– https://arxiv.org/abs/2302.10907.
Goodfellow I. et al. Generative adversarial nets // arXiv.– 2014.– https://arxiv.org/ abs/1406.2661.
Kingma D., Welling M. Auto-encoding variational bayes // arXiv.– 2013.– https:// arxiv.org/abs/1312.6114.
Rezende D., Mohamed Sh. Variational inference with normalizing flows // arXiv.– 2015. https://arxiv.org/abs/1505.05770.
Yang L. et al. Diffusion models: A comprehensive survey of methods and applications // arXiv.– 2022.– https://arxiv.org/abs/2209.00796.
Ho J., Jain A., Abbeel P. Denoising diffusion probabilistic models // Advances in neural information processing systems.– 2020.– Vol. 33.– P. 6840-6851.
Song Y., Ermon S. Generative modeling by estimating gradients of the data distribution // arXiv.– 2019.– https://arxiv.org/abs/1907.05600.
Song Y. et al. Score based generative modeling through stochastic differential equations // arXiv.– 2020.– https://arxiv.org/abs/2011.13456.
Austin J. Structured denoising diffusion models in discrete state-spaces // Advances in Neural Information Processing Systems.– 2021.– Vol. 34.– P. 17981-17993
Satorras V.G., Hoogeboom E., Welling M. E(n) Equivariant Graph Neural Networks // arXiv.– 2021.– https://arxiv.org/abs/2102.09844.
Tang X. et al. Mollm: A unified language model for integrating biomedical text with 2d and 3d molecular representations // bioRxiv.– 2023.– doi: http://dx.doi.org/10.1101/ 2023.11.25.568656.
Ramakrishnan R. et al. Quantum chemistry structures and properties of 134 kilo molecules // Scientific data.– 2014.– No. 1(1).– P. 1-7.
Axelrod S., Gomez-Bombarelli R. Geom, energy-annotated molecular conformations for property prediction and molecular generation // Scientific Data.– 2022.– No. 9(1). P. 185
Bickerton G.R. et al. Quantifying the chemical beauty of drugs // Nature chemistry. 2012.– No. 4(2).– P. 90–98.
G´omez-Bombarelli R. et al. Automatic chemical design using a data-driven continuous representation of molecules // ACS central science.– 2018.– No. 4(2).– P. 268–276.
Jin W., Barzilay R., Jaakkola T. Junction tree variational autoencoder for molecular graph generation // arXiv.– 2018.– https://arxiv.org/abs/1802.04364.
Satorras V.G. et al. E(n) Equivariant normalizing flows // arXiv.– 2021.– https://arxiv. org/abs/2105.09016.
Gebauer N., Gastegger M., Schutt K. Symmetry-adapted generation of 3d point sets for the targeted discovery of molecules // arXiv.– 2019.– https://arxiv.org/abs/1906.00957.
Hoogeboom E. et al. Equivariant diffusion for molecule generation in 3d // arXiv.– 2022. https://arxiv.org/abs/2203.17003.
Alex Morehead, Jianlin Cheng Geometry-complete diffusion for 3d molecule generation // arXiv.– 2023.– https://arxiv.org/abs/2302.04313.
Huang L. et al. Mdm: Molecular diffusion model for 3d molecule generation // arXiv. 2022.– https://arxiv.org/abs/2209.05710.
Xu M. et al. Geometric latent diffusion models for 3d molecule generation // arXiv.– 2023. https://arxiv.org/abs/2305.01140.
Huang H. et al. Learning joint 2d & 3d diffusion models for complete molecule generation // arXiv.– 2023.– https://arxiv.org/abs/2305.12347.
Vignac C. et al. Midi: Mixed graph and 3d denoising diffusion for molecule generation // arXiv.– 2023.– https://arxiv.org/abs/2302.09048.
Zubatiuk T., Isayev O. Development of multimodal machine learning potentials: toward a physics-aware artificial intelligence // Acc. Chem. Res. .– 2021.– Vol. 54.– P. 1575-1585.
Behler J. Four generations of high-dimensional neural network potentials // Chem. Rev. . 2021.– Vol. 121.– P. 10037-10072.
Smith J.S. et al. Less is more: sampling chemical space with active learning // J. Chem. Phys.– 2018.– Vol. 148.– 241733.
Smith J.S., Isayev O., Roitberg A.E. ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost // Chem. Sci.– 2017.– Vol. 8.– P. 3192 3203.
Zubatyuk R. et al. Accurate and transferable multitask prediction of chemical properties with an atoms-in-molecules neural network // ScienceAdvances .– 2019.– Vol. 5.– doi: http: //dx.doi.org/10.1126/sciadv.aav6490.
Li J. et al. Drug discovery approaches using quantum machine learning // arXiv.– 2021. https://arxiv.org/abs/2104.00746.
Romero J., Olson J.P., Aspuru-Guzik A. Quantum autoencoders for efficient compression of quantum data // Quantum Sci. Technol.– 2017.– Vol. 2.– 045001.
Xiang W. et al. Application of High Throughput Technologies in the Development of Acute Myeloid Leukemia Therapy: Challenges and Progress // Int. J. Mol. Sci..–2022.– Vol. 23.– 2863.– doi: http://dx.doi.org/10.3390/ijms23052863.
Cavasotto C.N. Binding free energy calculation using quantum mechanics aimed for drug lead optimization // Methods Mol. Biol.– 2020.– Vol. 2114.– P. 257-268.
Heinen, S. et al Predicting toxicity by quantum machine learning. // J. Phys. Commun. 2020.– Vol. 4.– 125012.
Outeiral, C. et al The prospects of quantum computing in computational molecular biology. // Wiley Interdiscip. Rev. Comput. Mol. Sci.– 2015.– Vol. 11.– e1481.
Jayatunga M.K.P. et al. AI in small-molecule drug discovery: a coming wave? // Nature Reviews Drug Discovery.– 2022.– Vol. 21.– P. 175-176.
Innovations in and around generative AI dominate and have transformative impact / Jackie Wiles.– 2022.– URL: https://shorturl.at/qBFN3.
