betaFeedback

WorksComputer Science

PRISM: Prior Relational Information for Self-supervised Modeling to Enhance Solubility OOD Generalization

Abstract

Solubility depends on intermolecular interactions between solutes and solvents, and is central to drug discovery, chemical and materials process engineering. Its prediction requires out-of-distribution (OOD) generalization to unseen solutes, solvents, and solvent mixtures, where supervised methods suffer from data scarcity, and existing self-supervised methods do not explicitly learn chemically meaningful relations. Learning chemically meaningful relational information without labels can alleviate data scarcity and improve OOD generalization, but remains challenging. We introduce Prior Relational Information for Self-supervised Modeling (PRISM) and apply it to solubility OOD generalization. PRISM is a label-free pretraining framework that converts expert-defined chemical priors into relational teachers. For solubility prediction, it trains four encoders to preserve the population-level relational information from priors capturing hydrophobic, electrostatic, polarizability, and substructural effects, using Kullback-Leibler (KL) divergence to match distance-based neighborhood distributions during pretraining. Across scaffold, solute, and solvent OOD settings, PRISM matches or surpasses supervised and self-supervised baselines. Under mixed-solvent OOD, freezing the four PRISM encoders and training a lightweight fusion-and-composition head outperforms strong baselines and substantially reduces run-to-run variation. Ablations and control studies confirm that gains stem from chemically meaningful relational structures baked into the encoders by PRISM. We expect this strategy of turning relational priors into self-supervision signals to extend beyond solubility prediction to domains where chemically or biologically meaningful relational priors are available.

Introduction

Solubility measures how much of a solute dissolves in a specified pure or mixed solvent under defined temperature and pressure, and it is a critical property across drug discovery, formulation, and pharmaceutical process development [1]. In practical applications, models often encounter unseen scaffolds, unseen solutes and solvents, and unseen binary solvent pairs in mixed-solvent systems, requiring solubility prediction beyond the training distribution. These shifts directly affect the solute–solvent and solvent–solvent interactions governing solubility [2, 3], making them natural out-of-distribution (OOD) scenarios for evaluating chemically meaningful molecular representations.

Recent solubility prediction methods model solvent context through recommendation, interaction-graph, hypergraph, or multicomponent architectures [4, 5, 6, 7]. Broader molecular relational learning methods explicitly model drug–drug or solute–solvent interactions [8, 9, 2], but usually require supervised pairs or task-specific data. Molecular self-supervised objectives based on augmentation-defined pairs, masking, or reconstruction [10, 11, 12, 13, 14] do not specify scientifically grounded population-level neighborhoods: which molecules should be close or far under a chemically meaningful relation. Classical manifold and molecular metric-learning methods can preserve predefined relations, but leave open how to transfer scientifically grounded population-level neighborhood geometry into molecular encoders without labels [15, 16]. These studies leave a research gap: label-free pretraining remains weakly aligned with the relational structure needed for solubility OOD generalization.

For solubility, a chemically grounded route can close this gap: by the empirical “like dissolves like” rule, solute–solvent compatibility is reflected by descriptor- or fingerprint-based physicochemical similarity [17, 18, 19]. We propose PRISM (Prior Relational Information for Self-supervised Modeling), which turns such similarity into a label-free relational teacher. PRISM uses these prior-defined relations to shape the geometry of learned molecular embeddings.

PRISM has two design choices: the prior source, which specifies the expert-defined relation, and the alignment design, which transfers that relation into embedding space. We use van der Waals surface area (VSA) descriptors [19] and Morgan fingerprints [18] as complementary physicochemical and substructural priors, and adopt distance-based distribution matching with prior-specific encoders fused only downstream (late fusion) as the alignment design among several candidates.

Our main results are OOD generalization of solubility prediction. PRISM outperforms supervised and SSL-pretrained baselines in cross-dataset generalization from BigSolDB to Leeds and SolProp for predicting the solubility of diverse solutes in pure solvents. With frozen prior-specific encoders, trainable fusion, and a composition-weighted head, it matches or improves end-to-end mixed-solvent baselines under solvent-pair-disjoint splits with better stability. Label energy (a form of graph Dirichlet energy) diagnostics and ablation controls further show that these gains come from prior–task compatibility, relational pretraining, and late fusion of prior-specific embeddings, rather than from descriptor inputs, reconstruction, architectural capacity, or coordinate-level imitation.

Our contributions are the following:

  1. We propose PRISM, a label-free molecular pretraining framework that bakes chemically meaningful prior knowledge into molecular representations through prior-induced relational modeling, providing a practical route to label-free relational representation learning.

  2. We build a PRISM-based solubility prediction model using VSA descriptors and Morgan fingerprints as complementary physicochemical and substructural priors. The model includes two stages: the pretraining of four prior-specific encoders and the weighted-sum late fusion of prior-specific embeddings during downstream tasks.

  3. We show that PRISM improves solubility-centered OOD generalization from scaffold splits to pure- and mixed-solvent settings, matching or surpassing strong baselines and improving stability under mixed-solvent OOD. Ablations and label-energy diagnostics support relational pretraining, prior complementarity, and prior–task compatibility.

Methodology

PRISM treats each chemical prior as a relational teacher, defining molecular closeness under a specific physicochemical criterion. We adopt D-MPNN as a strong 2D graph backbone [20]; the framework itself remains encoder-agnostic. We first discuss which prior relations to choose in § 2.1, and then describe how they are aligned in the embedding space and fused in § 2.2 and § 2.3.

Solubility-grounded prior spaces

The chemistry of solubility guides which prior relations PRISM uses. Solubility depends on solute–solvent interactions and molecular structure. Following the empirical “like dissolves like” principle, solubility is favored when solute and solvent share similar interaction components, including hydrophobic, electrostatic, and dispersion-related effects, or similar substructures capturing interactions not accounted for by those effects. PRISM therefore uses four complementary priors: SlogP_VSA for hydrophobic/lipophilic effects, SMR_VSA for polarizability/dispersion-relevant effects, PEOE_VSA for electrostatic effects, and Morgan fingerprints for substructure and functional-group patterns. Detailed descriptions of these prior spaces and other prior spaces are provided in Appendix D.

Prior-induced relational alignment

Let Dpre={xi}i=1N\mathcal{D}_{\mathrm{pre}}=\{x_i\}_{i=1}^{N} be an unlabeled set of molecular graphs and let S={sk}k=1K\mathcal{S}=\{s_k\}_{k=1}^{K} be a set of prior spaces. For each prior space sk∈Ss_k\in\mathcal{S}, a fixed prior function maps molecule xix_i to πisk\pi_i^{s_k}, and a prior-space comparator dskd^{s_k} measures pairwise relations. A prior-specific encoder GθkskG_{\theta_k}^{s_k} with parameters θk\theta_k maps the same molecule to an embedding riskr_i^{s_k}. PRISM aligns the relation induced by {risk}i=1N\{r_i^{s_k}\}_{i=1}^{N} with the relation induced by {πisk}i=1N\{\pi_i^{s_k}\}_{i=1}^{N} in a minibatch B\mathcal{B}. The alignment procedure is illustrated in Figure 1a.

Figure 1. PRISM overview. (a) Prior-guided alignment pretrains an encoder per prior space. (b) Weighted fusion produces a fused molecular embedding.

Concretely, two relational constructors Apriorsk\mathcal{A}_{\mathrm{prior}}^{s_k} and Aembsk\mathcal{A}_{\mathrm{emb}}^{s_k} map pairwise comparisons in B\mathcal{B} to two relational objects Rpriorsk\mathcal{R}_{\mathrm{prior}}^{s_k} and Rembsk\mathcal{R}_{\mathrm{emb}}^{s_k}:

Rpriorsk=Apriorsk({dsk(πisk,πjsk)}i,j∈B),Rembsk=Aembsk({dsk(risk,rjsk)}i,j∈B).(1)\mathcal{R}_{\mathrm{prior}}^{s_k}=\mathcal{A}_{\mathrm{prior}}^{s_k}\left(\left\{d^{s_k}\left(\pi_i^{s_k},\pi_j^{s_k}\right)\right\}_{i,j\in\mathcal{B}}\right),\qquad \mathcal{R}_{\mathrm{emb}}^{s_k}=\mathcal{A}_{\mathrm{emb}}^{s_k}\left(\left\{d^{s_k}\left(r_i^{s_k},r_j^{s_k}\right)\right\}_{i,j\in\mathcal{B}}\right). \tag*{(1)}

The objective of PRISM is to align the two relational objects, Rpriorsk\mathcal{R}_{\mathrm{prior}}^{s_k} and Rembsk\mathcal{R}_{\mathrm{emb}}^{s_k}, using a suitable discrepancy metric as the loss function:

LPRISMsk=Msk(Rpriorsk,Rembsk).(2)\mathcal{L}_{\mathrm{PRISM}}^{s_k}=\mathcal{M}^{s_k}\left(\mathcal{R}_{\mathrm{prior}}^{s_k},\mathcal{R}_{\mathrm{emb}}^{s_k}\right). \tag*{(2)}

Table 1 classifies alignment objectives by the relational objects they preserve. As a concrete example, pairwise distances can be converted into anchor-wise stochastic neighborhood distributions. For each prior space sks_k, anchor molecule xix_i, and candidate set Bi=B∖{xi}\mathcal{B}_i=\mathcal{B}\setminus\{x_i\}, RiskR_i^{s_k} denotes the prior-space probability distribution of xix_i, while PiskP_i^{s_k} denotes the corresponding embedding-space probability distribution. The relational objects are sets of anchor-wise distributions

Relational objectObjectiveDiscrepancy metricRelated instances
Pairwise distance matrixPointwise metric matchingDistance regressionTalibart et al. [21]
Neighbor orderingRank preservationRank / ListMLE lossPH-TDL [22]
Top-nn prior-space kNN graphNeighborhood-graph alignmentInfoNCE over graph positivesMoleco [23]
Pair-wise distance-based distributionDistribution matchingKL divergenceSNE-style [24] / GraphMSL [25]

Table 1. Relational objects and alignment objectives in PRISM with related instances.

Rpriorsk={Risk}xi∈B,Rembsk={Pisk}xi∈B.\mathcal{R}_{\mathrm{prior}}^{s_k}=\{R_i^{s_k}\}_{x_i\in\mathcal{B}},\qquad \mathcal{R}_{\mathrm{emb}}^{s_k}=\{P_i^{s_k}\}_{x_i\in\mathcal{B}}.

The probabilities in these two distributions are defined by

Risk(xj)=exp⁡(−βkdsk(πisk,πjsk))∑xl∈Biexp⁡(−βkdsk(πisk,πlsk)),Pisk(xj)=exp⁡(−βτδ(risk,rjsk))∑xl∈Biexp⁡(−βτδ(risk,rlsk)).(3)R_i^{s_k}(x_j) = \frac{\exp\left(-\beta_k d^{s_k}\left(\pi_i^{s_k},\pi_j^{s_k}\right)\right)}{\sum_{x_l \in B_i} \exp\left(-\beta_k d^{s_k}\left(\pi_i^{s_k},\pi_l^{s_k}\right)\right)}, \qquad P_i^{s_k}(x_j) = \frac{\exp\left(-\beta_\tau\delta\left(r_i^{s_k},r_j^{s_k}\right)\right)}{\sum_{x_l \in B_i} \exp\left(-\beta_\tau\delta\left(r_i^{s_k},r_l^{s_k}\right)\right)}. \tag*{(3)}

Here dskd^{s_k} is the prior-space distance, δ(risk,rjsk)=1−cos⁡(risk,rjsk)\delta\left(r_i^{s_k},r_j^{s_k}\right) = 1 - \cos\left(r_i^{s_k},r_j^{s_k}\right), and βk,βτ\beta_k,\beta_\tau control locality. The loss function aggregates anchor-wise KL divergence over the minibatch B\mathcal{B}:

LKLsk=1∣B∣∑xi∈BKL(Risk∥Pisk).(4)\mathcal{L}_{\mathrm{KL}}^{s_k} = \frac{1}{\lvert\mathcal{B}\rvert} \sum_{x_i \in\mathcal{B}} \mathrm{KL}\left(R_i^{s_k} \mathbin{\|} P_i^{s_k}\right). \tag*{(4)}

Because Eq. (3) normalizes each anchor separately, these neighborhood distributions are conditional and generally asymmetric, retaining anchor-specific neighborhood geometry. Section 3.2 evaluates the four alignment objectives and explains why distribution matching is used in PRISM. Appendix § 12.§ 2 provides additional theoretical analysis on these alignment objectives and compares the conditional formulation with a symmetric SNE-style alternative.

Weighted-sum fusion

References

  1. [1]Luigi Da Vià, Matthias Depoortere, Robert D Willacy, Alastair J Roberts, Pandian Sokkar, Mathieu Fossépré, Andrew Ruba, and Magdalena A Zwierzyna. Enabling data-driven solubility modeling at gsk: Enhancing pure predictions for mutagenic impurities. Organic Process Research & Development, 28(12):4215–4224, 2024.DOI
  2. [2]Yeyun Chen. Relational invariant learning for robust solvation free energy prediction. In International Conference on Machine Learning, pages 72210–72227. PMLR, 2025.
  3. [3]Jianyuan Deng, Zhibo Yang, Hehe Wang, Iwao Ojima, Dimitris Samaras, and Fusheng Wang. A systematic study of key elements underlying molecular property prediction. Nature Communications, 14(1):6395, 2023.DOI
  4. [4]Jan Wollschläger and Floriane Montanari. Solvmate – a hybrid physical/ML approach to solvent recommendation leveraging a rank-based problem framework. Digital Discovery, 3:1749–1760, 2024. doi: 10.1039/D4DD00138A.DOI
  5. [5]Vansh Ramani and Tarak Karmakar. Graph neural networks for predicting solubility in diverse solvents using MolMerger incorporating solute–solvent interactions. Journal of Chemical Theory and Computation, 20(15):6549–6558, 2024. doi: 10.1021/acs.jctc.4c00382.DOI
  6. [6]Wenjie Du, Shuai Zhang, Zhaohui Cai, Xuqiang Li, Zhiyuan Liu, Junfeng Fang, Jianmin Wang, Xiang Wang, and Yang Wang. Molecular merged hypergraph neural network for explainable solvation Gibbs free energy prediction. Research, 8:0740, 2025. doi: 10.34133/research.0740.DOI
  7. [7]Hojin Jung, Christopher D. Stubbs, Sabari Kumar, Raúl Pérez-Soto, Su-min Song, Yeonjoon Kim, and Seonah Kim. Enhancing predictive models for solubility in multicomponent solvent systems using semi-supervised graph neural networks. Digital Discovery, 4:1492–1504, 2025. doi: 10.1039/D5DD00015G.DOI
  8. [8]Namkyeong Lee, Dongmin Hyun, Gyoung S Na, Sungwon Kim, Junseok Lee, and Chanyoung Park. Conditional graph information bottleneck for molecular relational learning. In International Conference on Machine Learning, pages 18852–18871. PMLR, 2023.arxiv.org/abs/2305.01520
  9. [9]Namkyeong Lee, Kanghoon Yoon, Gyoung S. Na, Sein Kim, and Chanyoung Park. Shift-robust molecular relational learning with causal substructure. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 1200–1212. Association for Computing Machinery, 2023. doi: 10.1145/3580305.3599437.DOI
  10. [10]Y. Wang, J. Wang, Z. Cao, and A. Barati Farimani. Molecular contrastive learning of representations via graph neural networks. Nature Machine Intelligence, 4:279–287, 2022. doi: 10.1038/s42256-022-00447-x. URL https://www.nature.com/articles/s42256-022-00447-x.DOI
  11. [11]Tianyang Hu, Zhili Liu, Fengwei Zhou, Wenjia Wang, and Weiran Huang. Your contrastive learning is secretly doing stochastic neighbor embedding. arXiv preprint arXiv:2205.14814, 2022.arxiv.org/abs/2205.14814
  12. [12]Gabriel A Pinheiro, Juarez LF Da Silva, and Marcos G Quiles. Smiclr: contrastive learning on multiple molecular representations for semisupervised and unsupervised representation learning. Journal of Chemical Information and Modeling, 62(17):3948–3960, 2022.DOI
  13. [13]Ross Irwin, Spyridon Dimitriadis, Jiazhen He, and Esben Jannik Bjerrum. Chemformer: a pretrained transformer for computational chemistry. Machine Learning: Science and Technology, 3(1):015022, 2022.
  14. [14]Pascal Vincent. A connection between score matching and denoising autoencoders. Neural computation, 23(7):1661–1674, 2011.DOI
  15. [15]Michele Ceriotti, Gareth A. Tribello, and Michele Parinello. Simplifying the representation of complex free-energy landscapes using sketch-map. Proceedings of the National Academy of Sciences, 108(32):13023–13028, 2011. doi: 10.1073/pnas.1108486108.DOI
  16. [16]Damien E. Coupry and Peter Pogány. Application of deep metric learning to molecular graph similarity. Journal of Cheminformatics, 14:11, 2022. doi: 10.1186/s13321-022-00595-7.DOI
  17. [17]Roberto Todeschini and Viviana Consonni. Molecular descriptors for chemoinformatics: volume I: alphabetical listing/volume II: appendices, references. John Wiley & Sons, 2009.
  18. [18]David Rogers and Mathew Hahn. Extended-connectivity fingerprints. Journal of chemical information and modeling, 50(5):742–754, 2010.DOI
  19. [19]Paul Labute. A widely applicable set of descriptors. Journal of Molecular Graphics and Modelling, 18(4-5):464–477, 2000.pubmed.ncbi.nlm.nih.gov/11143563
  20. [20]Kevin Yang, Kyle Swanson, Wengong Jin, Connor Coley, Philipp Eiden, Hua Gao, Angel Guzman-Perez, Timothy Hopper, Brian Kelley, Miriam Mathea, Andrew Palmer, Volker Settels, Tommi Jaakkola, Klavs Jensen, and Regina Barzilay. Analyzing learned molecular representations for property prediction. Journal of Chemical Information and Modeling, 59(8):3370–3388, 2019. doi: 10.1021/acs.jcim.9b00237.
  21. [21]Hugo Talibart and Dimitri Gilis. Learning a chemistry-aware latent space for molecular encoding and generation with a large-scale transformer variational autoencoder. Journal of Cheminformatics, 18(1):123, 2026. doi: 10.1186/s13321-026-01243-0.
  22. [22]Yuankai Luo, Lei Shi, and Veronika Thost. Improving self-supervised molecular representation learning using persistent homology. Advances in Neural Information Processing Systems, 36: 34043–34073, 2023.
  23. [23]Jun-Hyung Park, Hyuntae Park, Yeachan Kim, Woosang Lim, and SangKeun Lee. Moleco: Molecular contrastive learning with chemical language models for molecular property prediction. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 408–420, 2024.
  24. [24]Geoffrey E Hinton and Sam Roweis. Stochastic neighbor embedding. Advances in neural information processing systems, 15, 2002.
  25. [25]Hao Xu, Zhengyang Zhou, and Pengyu Hong. Graph multi-similarity learning for molecular property prediction. arXiv preprint arXiv:2401.17615, 2024.
  26. [26]Zeqing Bao, Gary Tom, Austin Cheng, Jeffrey Watchorn, Alán Aspuru-Guzik, and Christine Allen. Towards the prediction of drug solubility in binary solvent mixtures at various temperatures using machine learning. Journal of Cheminformatics, 16(1):117, 2024.
  27. [27]Lucas Attia, Jackson W Burns, Patrick S Doyle, and William H Green. Data-driven organic solubility prediction at the limit of aleatoric uncertainty. Nature Communications, 16(1):7497, 2025.
  28. [28]Laurens Van Der Maaten and Kilian Weinberger. Stochastic triplet embedding. In 2012 IEEE International Workshop on Machine Learning for Signal Processing, pages 1–6. IEEE, 2012.DOI
  29. [29]Eric Inae, Gang Liu, and Meng Jiang. Motif-aware attribute masking for molecular graph pre-training. arXiv preprint arXiv:2309.04589, 2023.
  30. [30]Yashaswi Pathak, Siddhartha Laghuvarapu, Sarvesh Mehta, and U. Deva Priyakumar. Chemically interpretable graph interaction network for prediction of pharmacokinetic properties of drug-like molecules. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 873–880, 2020. doi: 10.1609/aaai.v34i01.5433.
  31. [31]Wenjie Du, Shuai Zhang, Di Wu, Jun Xia, Ziyuan Zhao, Junfeng Fang, and Yang Wang. MMGNN: A molecular merged graph neural network for explainable solvation free energy prediction. In Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, pages 5808–5816, 2024. doi: 10.24963/ijcai.2024/642.
  32. [32]Walid Ahmad, Elana Simon, Seyone Chithrananda, Gabriel Grand, and Bharath Ramsundar. Chemberta-2: Towards chemical foundation models, 2022.arxiv.org/abs/2209.01712
  33. [33]Yu Rong, Yatao Bian, Tingyang Xu, Weiyang Xie, Ying Wei, Wenbing Huang, and Junzhou Huang. Self-supervised graph transformer on large-scale molecular data. Advances in neural information processing systems, 33:12559–12571, 2020.
  34. [34]Francesca Cenci, Samir Diab, Paola Ferrini, Cataijna Harabajiu, Massimiliano Barolo, Fabrizio Bezzo, and Pierantonio Facco. Predicting drug solubility in organic solvents mixtures: A machine-learning approach supported by high-throughput experimentation. International Journal of Pharmaceutics, 660:124233, 2024. doi: 10.1016/j.ijpharm.2024.124233.DOI
  35. [35]Dennis A Dougherty. The cation- π interaction in chemistry and biology. Chemical Reviews, 125(5):2793–2808, 2025.DOI
  36. [36]Boshra Ariguib, Mathias Niepert, and Andrei Manolache. Learning the neighborhood: Contrast-free self-supervised molecular graph pretraining. In New Perspectives in Graph Machine Learning, 2025.arxiv.org/abs/2509.22468
  37. [37]Kisung Moon, Hyeon-Jin Im, and Sunyoung Kwon. 3d graph contrastive learning for molecular property prediction. Bioinformatics, 39(6):btad371, 2023.DOI
  38. [38]Jiying Zhang, Zijing Liu, Yu Wang, Bin Feng, and Yu Li. Subgdiff: A subgraph diffusion model to improve molecular representation learning. Advances in Neural Information Processing Systems, 37:29620–29656, 2024.arxiv.org/abs/2405.05665
  39. [39]Zewei Ji, Runhan Shi, Jiarui Lu, Fang Li, and Yang Yang. Relmole: Molecular representation learning based on two-level graph similarities. Journal of Chemical Information and Modeling, 62(22):5361–5372, 11 2022. ISSN 1549-9596. doi: 10.1021/acs.jcim.2c00798. URL https://doi.org/10.1021/acs.jcim.2c00798.
  40. [40]Xiaohong Ji, Zhen Wang, Zhifeng Gao, Hang Zheng, Linfeng Zhang, Guolin Ke, and Weinan E. Exploring molecular pretraining model at scale. volume 37, pages 46956–46978, 2024.DOI
  41. [41]Oscar Méndez-Lucio, Christos A Nicolaou, and Berton Earnshaw. Mole: a foundation model for molecular graphs using disentangled attention. Nature Communications, 15(1):9431, 2024.
  42. [42]Yi Zhong, Gaozheng Li, Ji Yang, Houbing Zheng, Yongqiang Yu, Jiheng Zhang, Heng Luo, Biao Wang, and Zuquan Weng. Learning motif-based graphs for drug–drug interaction prediction via local–global self-attention. Nature Machine Intelligence, 6(9):1094–1105, 2024.DOI
  43. [43]Wenbin Ye and Quan Qian. Forecasting the molecular interactions: A hypergraph-based neural network for molecular relational learning. Knowledge-Based Systems, 300:112177, 2024.
  44. [44]Ella Miray Rajaonson, Mahyar Rajabi Kochi, Luis Martin Mejía Mendoza, Seyed Mohamad Moosavi, and Benjamin Sanchez-Lengeling. Chemixhub: Datasets and benchmarks for chemical mixture property prediction. arXiv preprint arXiv:2506.12231, 2025.arxiv.org/abs/2506.12231
  45. [45]Charles M Hansen. Hansen solubility parameters: a user’s handbook. CRC press, 2007.
  46. [46]Dávid Bajusz, Anita Rácz, and Károly Héberger. Why is Tanimoto index an appropriate choice for fingerprint-based similarity calculations? Journal of Cheminformatics, 7(1):20, 2015. doi: 10.1186/s13321-015-0069-3.DOI
  47. [47]Connor W Coley. Defining and exploring chemical spaces. Trends in Chemistry, 3(2):133–145, 2021.DOI
  48. [48]Yu Liu, Cole Mathis, Michał Dariusz Bajczyk, Stuart M Marshall, Liam Wilbraham, and Leroy Cronin. Exploring and mapping chemical space with molecular assembly trees. Science Advances, 7(39):eabj2465, 2021.DOI
  49. [49]Alexios Koutsoukas, Shardul Paricharak, Warren RJD Galloway, David R Spring, Adriaan P IJzerman, Robert C Glen, David Marcus, and Andreas Bender. How diverse are diversity assessment methods? a comparative analysis and benchmarking of molecular descriptor space. Journal of chemical information and modeling, 54(1):230–242, 2014.DOI
  50. [50]Andreas Bender, Jeremy L Jenkins, Josef Scheiber, Sai Chetan K Sukuru, Meir Glick, and John W Davies. How similar are similarity searching methods? a principal component analysis of molecular descriptor space. Journal of chemical information and modeling, 49(1):108–119, 2009.DOI
  51. [51]Florence H Vermeire, Yunsie Chung, and William H Green. Predicting solubility limits of organic solutes for a wide range of solvents and temperatures. Journal of the American Chemical Society, 144(24):10785–10797, 2022.DOI
  52. [52]Samuel Boobier, David RJ Hose, A John Blacker, and Bao N Nguyen. Machine learning with physicochemical relationships: solubility prediction in organic solvents and water. Nature communications, 11(1):5753, 2020.DOI
  53. [53]Jeremy R. Ash, Cas Wognum, Raquel Rodríguez-Pérez, Matteo Aldeghi, Alan C. Cheng, Djork-Arné Clevert, Ola Engkvist, Cheng Fang, Daniel J. Price, Jacqueline M. Hughes-Oliver, and W. Patrick Walters. Practically significant method comparison protocols for machine learning in small molecule drug discovery. Journal of Chemical Information and Modeling, 65(18):9398–9411, 2025. doi: 10.1021/acs.jcim.5c01609.DOI
  54. [54]Henrike Veith, Noel Southall, Ruili Huang, Tim James, Darren Fayne, Natalia Artemenko, Min Shen, James Inglese, Christopher P Austin, David G Lloyd, et al. Comprehensive characterization of cytochrome p450 isozyme selectivity across chemical libraries. Nature biotechnology, 27(11):1050–1055, 2009.DOI
  55. [55]Yin Fang, Qiang Zhang, Ningyu Zhang, Zhuo Chen, Xiang Zhuang, Xin Shao, Xiaohui Fan, and Huajun Chen. Knowledge graph-enhanced molecular contrastive learning with functional prompt. Nature Machine Intelligence, 5:542–553, May 2023. ISSN 2522-5839. URL https://www.nature.com/articles/s42256-023-00654-0.DOI

Paper details

vlnc:W_yqq5ss5n
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Computer Science

Not recorded for this paper

View statement
Looking for related papers…
Contents