Actively Searching: Inverse Design of Novel Molecules with Simultaneously Optimized Properties.

Combining quantum chemistry characterizations with generative machine learning models has the potential to accelerate molecular discovery. In this paradigm, quantum chemistry acts as a relatively cost-effective oracle for evaluating the properties of particular molecules, while generative models provide a means of sampling chemical space based on learned structure-function relationships. For practical applications, multiple potentially orthogonal properties must be optimized in tandem during a discovery workflow. This carries additional difficulties associated with the specificity of the targets and the ability for the model to reconcile all properties simultaneously. Here, we demonstrate an active learning approach to improve the performance of multi-target generative chemical models. We first demonstrate the effectiveness of a set of baseline models trained on single property prediction tasks in generating novel compounds (i.e., not present in the training data) with various property targets, including both interpolative and extrapolative generation scenarios. For property ranges where accurate targeting proves difficult, the novel compounds suggested by the model are characterized using quantum chemistry and the new molecules closest to expressing the desired properties are fed back into the generative model for additional training. This gradually improves the generative models' understanding of targeted areas of chemical space and shifts the distribution of the generated compounds toward the targeted values. We then demonstrate the effectiveness of this active learning approach in generating compounds with multiple chemical constraints, including vertical ionization potential, electron affinity, and dipole moment targets, and validate the results at the ωB97X-D3/def2-TZVP level. This method requires no modifications to extant generative approaches, but rather utilizes their inherent generative and predictive aspects for self-refinement, and can be applied to situations where any number of properties with varying degrees of correlation must be optimized simultaneously.

Transferable Ring Corrections for Predicting Enthalpy of Formation of Cyclic Compounds.

Computational predictions of the thermodynamic properties of molecules and materials play a central role in contemporary reaction prediction and kinetic modeling. Due to the lack of experimental data and computational cost of high-level quantum chemistry methods, approximate methods based on additivity schemes and more recently machine learning are currently the only approaches capable of supplying the chemical coverage and throughput necessary for such applications. For both approaches, ring-containing molecules pose a challenge to transferability due to the nonlocal interactions associated with conjugation and strain that significantly impact thermodynamic properties. Here, we report the development of a self-consistent approach for parameterizing transferable ring corrections based on high-level quantum chemistry. The method is benchmarked against both the Pedley-Naylor-Kline experimental dataset for C-, H-, O-, N-, S-, and halogen-containing cyclic molecules and a dataset of Gaussian-4 quantum chemistry calculations. The prescribed approach is demonstrated to be superior to existing ring corrections while maintaining extensibility to arbitrary chemistries. We have also compared this ring-correction scheme against a novel machine learning approach and demonstrate that the latter is capable of exceeding the performance of physics-based ring corrections.

Simpler is Better: How Linear Prediction Tasks Improve Transfer Learning in Chemical Autoencoders.

Transfer learning is a subfield of machine learning that leverages proficiency in one or more prediction tasks to improve proficiency in a related task. For chemical property prediction, transfer learning models represent a promising approach for addressing the data scarcity limitations of many properties by utilizing potentially abundant data from one or more adjacent applications. Transfer learning models typically utilize a latent variable that is common to several prediction tasks and provides a mechanism for information exchange between tasks. For chemical applications, it is still largely unknown how correlation between the prediction tasks affects performance, the limitations on the number of tasks that can be simultaneously trained in these models before incurring performance degradation, and if transfer learning positively or negatively affects ancillary model properties. Here we investigate these questions using an autoencoder latent space as a latent variable for transfer learning models for predicting properties from the QM9 data set that have been supplemented with semiempirical quantum chemistry calculations. We demonstrate that property prediction can be counterintuitively improved by utilizing a simpler linear predictor model, which has the effect of forcing the latent space to organize linearly with respect to each property. In data scarce prediction tasks, the transfer learning improvement is dramatic, whereas in data rich prediction tasks, there appears to be little adverse impact of transfer learning on prediction performance. The transfer learning approach demonstrated here thus represents a highly advantageous supplement to property prediction models with no downside in implementation.

Improved Chemical Prediction from Scarce Data Sets via Latent Space Enrichment.

Modern machine learning provides promising methods for accelerating the discovery and characterization of novel chemical species. However, in many areas experimental data remain costly and scarce, and computational models are unavailable for targeted figures of merit. Here we report a promising pathway to address this challenge by using chemical latent space enrichment, whereby disparate data sources are combined in joint prediction tasks to enable improved prediction in data-scarce applications. The approach is demonstrated for p Ka prediction of moderately sized molecular species using a combination of experimentally available p Ka data and density functional theory-based characterizations of the (de)protonation free energy. A novel autoencoder framework is used to create a continuous chemical latent space that is then used in single and joint training tasks for property prediction. By combining these two data sets in a jointly trained autoencoder framework, we observe mutual improvement in property prediction tasks in the scarce data limit. We also demonstrate an enrichment mechanism that is unique to latent space training, whereby training on excess computational data can mitigate the prediction losses associated with scarce experimental data and advantageously organize the latent space. These results demonstrate that disparate chemical data sources can be advantageously combined in an autoencoder framework with potential general application to data-scarce chemical learning tasks.

Sodium Alginate Toughening of Gelatin Hydrogels.

Gelatin is a popular material for the creation of tissue phantoms due to its ease-of-use, safety, low relative cost, and its amenability to tuning physical properties through the use of additives. One difficulty that arises when using gelatin, especially in low concentrations, is the brittleness of the material. In this paper, we show that small additions of another common biological polymer, sodium alginate, significantly increase the toughness of gelatin without changing the Young's modulus or other low-strain stress relaxation properties of the material. Samples were characterized using ramp-hold stress relaxation tests. The experimental data from these tests were then fit to the Generalized Maxwell (GM) model, as well as two models based on a fractional calculus approach: the Kelvin-Voigt Fractional Derivative (KVFD) and Fractional Maxwell (FM) models. We found that for our samples, the fractional models provided better fits with fewer parameters, and at strains within the linear elastic region, the linear viscoelastic parameters of the alginate/gelatin and pure gelatin samples were essentially indistinguishable. When the same ramp-hold stress relaxation experiments were run at high strains outside of the linear elastic region, we observed a shift in stress relaxation to shorter time scales with increasing sodium alginate addition, which may be associated with an increase in fluidity within the gelatin matrix. This leads us to believe that sodium alginate acts to enhance the viscosity within the fluidic region of the gelatin matrix, providing additional energy dissipation without raising the modulus of the material. These results are applicable to anyone desiring independent control of the Young's modulus and toughness in preparing tissue phantoms, and suggest that sodium alginate should be added to low-modulus gelatin for use in biological and medical testing applications.