1. the complete title of one (or more) paper(s) published in the open literature describing the work that the author claims describes a human-competitive result; Hybrid Generative AI for De Novo Design of Co-Crystals with Enhanced Tabletability ---------------------------------------------------------------------------------------------------- 2. the name, complete physical mailing address, e-mail address, and phone number of EACH author of EACH paper(s); Nina Gubina email: gubina@scamt-itmo.ru phone: +7999527398 ITMO University 49 Kronverksky Pr. St. Petersburg 197101 Russian Federation Andrei Dmitrenko email: dmitrenko@scamt-itmo.ru phone: +79213960809 ITMO University 49 Kronverksky Pr. St. Petersburg 197101 Russian Federation Gleb Solovev email: glebsolo46@gmail.com phone: +79373536281 ITMO University 49 Kronverksky Pr. St. Petersburg 197101 Russian Federation Lyubov Yamshchikova email: YamLyubov@gmail.com phone: +79046118839 ITMO University 49 Kronverksky Pr. St. Petersburg 197101 Russian Federation Oleg Petrov email: ogbellew@gmail.com phone: +7 931 209 2345 ITMO University 49 Kronverksky Pr. St. Petersburg 197101 Russian Federation Ivan Lebedev email: lis@isc-ras.ru phone: +79969196195 Ivanovo State University of Chemistry and Technology Sheremetevsky avenue, 7 Ivanovo 153000 Russian Federation Nikita Serov email: serov@scamt-itmo.ru phone: +79315877336 ITMO University 49 Kronverksky Pr. St. Petersburg 197101 Russian Federation Grigorii Kirgizov email: gkirgizov@yandex.ru phone: +79312911531 ITMO University 49 Kronverksky Pr. St. Petersburg 197101 Russian Federation Nikolay Nikitin email: nnikitin@itmo.ru phone: +79062434402 ITMO University 49 Kronverksky Pr. St. Petersburg 197101 Russian Federation Vladimir Vinogradov email: vinogradov@scamt-itmo.ru phone: +79218906773 ITMO University 49 Kronverksky Pr. St. Petersburg 197101 Russian Federation ---------------------------------------------------------------------------------------------------- 3. the name of the corresponding author (i.e., the author to whom notices will be sent concerning the competition); Nina Gubina ---------------------------------------------------------------------------------------------------- 4. the abstract of the paper(s); Co-crystallization is an accessible way to control physicochemical characteristics of organic crystals, which finds many biomedical applications. In this work, we present Generative Method for Co-crystal Design (GEMCODE), a novel pipeline for automated co-crystal screening based on the hybridization of deep generative models and evolutionary optimization for broader exploration of the target chemical space. GEMCODE enables fast de novo co-crystal design with target tabletability profiles, which is crucial for the development of pharmaceuticals. With a series of experimental studies highlighting validation and discovery cases, we show that GEMCODE is effective even under realistic computational constraints. Furthermore, we explore the potential of language models in generating co-crystals. Finally, we present numerous previously unknown co-crystals predicted by GEMCODE and discuss its potential in accelerating drug development. ---------------------------------------------------------------------------------------------------- 5. a list containing one or more of the eight letters (A, B, C, D, E, F, G, or H) that correspond to the criteria (see above) that the author claims that the work satisfies; (B) The result is equal to or better than a result that was accepted as a new scientific result at the time when it was published in a peer-reviewed scientific journal. (F) The result is equal to or better than a result that was considered an achievement in its field at the time it was first discovered. (G) The result solves a problem of indisputable difficulty in its field. ---------------------------------------------------------------------------------------------------- 6. a statement stating why the result satisfies the criteria that the contestant claims (see examples of statements of human-competitiveness as a guide to aid in constructing this part of the submission); The proposed approach to the co-crystals (crystalline solids featuring a drug molecule and a coformer) design makes it possible to identify the high variety of the coformers for different tasks. The approach is aimed at the creation of coformers with a high diversity using the combination of evolutionary optimisation and deep generative models. It makes it significantly different from existing drug design tools and allows obtaining human-competitive results in different tasks. The software implementation of the proposed approach is available in the GEMCODE (https://github.com/ai-chem/GEMCODE). (B) (The result is equal to or better than a result that was accepted as a new scientific result at the time when it was published in a peer-reviewed scientific journal.) Previous research has proposed AI-based systems to predict co-crystal formation probabilities [1-3] and properties [4, 5], aiding in co-crystal design. However, these efforts remain sporadic and do not substantially simplify the co-crystal design process. This study introduces an automated GEMCODE pipeline utilizing a hybrid evolutionary approach for co-crystal design. It efficiently identifies promising co-crystals with enhanced properties while minimizing experimental needs, offering a more effective and versatile solution than existing methods. (F) (The result is equal to or better than a result that was considered an achievement in its field at the time it was first discovered.) GEMCODE surpasses existing tools by fully automating the design of co-crystals. Previously, the process of selecting molecular pairs was conducted almost manually [6], occasionally utilizing isolated AI tools [1-5]. GEMCODE transforms this into a fully automated system. It innovatively combines evolutionary optimization with a generative approach, enabling the discovery of new coformers and expanding the chemical space beyond the established options previously accessible to experimental chemists. (G) (The result solves a problem of indisputable difficulty in its field.) Pharmaceutical co-crystals are increasingly utilized across industries, especially in pharmaceuticals, for their enhanced properties such as solubility and stability. Although they offer substantial benefits, the process of identifying suitable molecular combinations for property optimization remains challenging and often requires extensive experimental screening. The GEMCODE pipeline significantly impacts co-crystal design by automating and optimizing in silico selection of coformers, thus reducing resource requirements for experiments. During GEMCODE validation, we demonstrated its capability to identify co-crystals with improved tabletability for three drugs (Paracetamol, Nicorandil, Rivaroxaban). The proposed co-crystals were confirmed in the literature as having enhanced tabletability [7-10], showcasing the unique potential of GEMCODE to improve molecular properties. References: [1] Jiang, Y., Yang, Z., Guo, J., Li, H., Liu, Y., Guo, Y., ... & Pu, X. (2021). Coupling complementary strategy to flexible graph neural network for quick discovery of coformer in diverse co-crystal materials. Nature Communications, 12(1), 5950. [2] Ahmadi, S., Ghanavati, M. A., & Rohani, S. (2024). Machine learning-guided prediction of cocrystals using point cloud-based molecular representation. Chemistry of Materials, 36(3), 1153-1161. [3] Song, Y., Ding, Y., Su, J., Li, J., & Ji, Y. (2025). Unlocking the Potential of Machine Learning in Co‐crystal Prediction by a Novel Approach Integrating Molecular Thermodynamics. Angewandte Chemie International Edition, e202502410. [4] Gamidi, R. K., & Rasmuson, Å. C. (2020). Analysis and artificial neural network prediction of melting properties and ideal mole fraction solubility of cocrystals. Crystal Growth & Design, 20(9), 5745-5759. [5] Guo, J., Sun, M., Zhao, X., Shi, C., Su, H., Guo, Y., & Pu, X. (2023). General graph neural network-based model to accurately predict cocrystal density and insight from data quality and feature representation. Journal of Chemical Information and Modeling, 63(4), 1143-1156. [6] Khudaida, S. H., Yen, Y. T., & Su, C. S. (2024). Cocrystal screening of anticancer drug p-toluenesulfonamide and preparation by supercritical antisolvent process. The Journal of Supercritical Fluids, 204, 106106. [7] Mannava, M. C., Gunnam, A., Lodagekar, A., Shastri, N. R., Nangia, A. K., & Solomon, K. A. (2021). Enhanced solubility, permeability, and tabletability of nicorandil by salt and cocrystal formation. CrystEngComm, 23(1), 227-237. [8] Kale, D. P., Puri, V., Kumar, A., Kumar, N., & Bansal, A. K. (2020). The role of cocrystallization-mediated altered crystallographic properties on the tabletability of rivaroxaban and malonic acid. Pharmaceutics, 12(6), 546. [9] Karki, S., Friščić, T., Fábián, L., Laity, P. R., Day, G. M., & Jones, W. (2009). Improving mechanical properties of crystalline solids by cocrystal formation: new compressible forms of paracetamol. Advanced materials, 21(38‐39), 3905-3909. [10] Maeno, Y., Fukami, T., Kawahata, M., Yamaguchi, K., Tagami, T., Ozeki, T., ... & Tomono, K. (2014). Novel pharmaceutical cocrystal consisting of paracetamol and trimethylglycine, a new promising cocrystal former. International journal of pharmaceutics, 473(1-2), 179-186. ---------------------------------------------------------------------------------------------------- 7. a full citation of the paper (that is, author names; title, publication date; name of journal, conference, or book in which article appeared; name of editors, if applicable, of the journal or edited book; publisher name; publisher city; page numbers, if applicable); Nina Gubina, Andrei Dmitrenko, Gleb Solovev, Lyubov Yamshchikova, Oleg Petrov, Ivan Lebedev, Nikita Serov, Grigorii Kirgizov, Nikolay Nikitin, Vladimir Vinogradov. Hybrid Generative AI for De Novo Design of Co-Crystals with Enhanced Tabletability // The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024 (https://openreview.net/forum?id=G4vFNmraxj). ---------------------------------------------------------------------------------------------------- 8. a statement either that "any prize money, if any, is to be divided equally among the co-authors" OR a specific percentage breakdown as to how the prize money, if any, is to be divided among the co-authors; Any prize money, if any, is to be divided equally among all co-authors. ---------------------------------------------------------------------------------------------------- 9. a statement stating why the authors expect that their entry would be the "best" The GEMCODE approach is now open-source on GitHub (https://github.com/ai-chem/GEMCODE), providing global access to researchers. This work is unique in its automated co-crystal design, utilizing a novel combination of generative AI and evolutionary optimization. The GEMCODE pipeline enhances the efficiency of co-crystal identification and increases the chemical diversity available for exploration. Our method demonstrates superior performance compared to human design in terms of resource efficiency, while also automating the process to achieve advanced results for various target molecules automatically. ---------------------------------------------------------------------------------------------------- 10. An indication of the general type of genetic or evolutionary computation used, such as GA (genetic algorithms), GP (genetic programming), ES (evolution strategies), EP (evolutionary programming), LCS (learning classifier systems), GI (genetic improvement), GE (grammatical evolution), GEP (gene expression programming), DE (differential evolution), etc. GP (genetic programming) ---------------------------------------------------------------------------------------------------- 11. The date of publication of each paper. If the date of publication is not on or before the deadline for submission, but instead, the paper has been unconditionally accepted for publication and is "in press" by the deadline for this competition, the entry must include a copy of the documentation establishing that the paper meets the "in press" requirement. 2024, December