Advancing material property prediction: using physics-informed machine learning models for viscosity.

In materials science, accurately computing properties like viscosity, melting point, and glass transition temperatures solely through physics-based models is challenging. Data-driven machine learning (ML) also poses challenges in constructing ML models, especially in the material science domain where data is limited. To address this, we integrate physics-informed descriptors from molecular dynamics (MD) simulations to enhance the accuracy and interpretability of ML models. Our current study focuses on accurately predicting viscosity in liquid systems using MD descriptors. In this work, we curated a comprehensive dataset of over 4000 small organic molecules' viscosities from scientific literature, publications, and online databases. This dataset enabled us to develop quantitative structure-property relationships (QSPR) consisting of descriptor-based and graph neural network models to predict temperature-dependent viscosities for a wide range of viscosities. The QSPR models reveal that including MD descriptors improves the prediction of experimental viscosities, particularly at the small data set scale of fewer than a thousand data points. Furthermore, feature importance tools reveal that intermolecular interactions captured by MD descriptors are most important for viscosity predictions. Finally, the QSPR models can accurately capture the inverse relationship between viscosity and temperature for six battery-relevant solvents, some of which were not included in the original data set. Our research highlights the effectiveness of incorporating MD descriptors into QSPR models, which leads to improved accuracy for properties that are difficult to predict when using physics-based models alone or when limited data is available.

The CCR2/MCP-1 Chemokine Pathway and Lung Adenocarcinoma.

Host anti-tumor immunity can be hindered by various mechanisms present within the tumor microenvironment, including the actions of myeloid-derived suppressor cells (MDSCs). We investigated the role of the CCR2/MCP-1 pathway in MDSC-associated tumor progression in murine lung cancer models. Phenotypic profiling revealed maximal expression of CCR2 by tumor-resident MDSCs, and MCP-1 by transplanted TC1 tumor cells, respectively. Use of CCR2-knockout (CCR2-KO) mice showed dependence of tumor growth on CCR2 signaling. Tumors in CCR2-KO mice had fewer CCR2low MDSCs, CD4 T cells and Tregs than WT mice, and increased infiltration by CD8 T cells producing IFN-γ and granzyme-B. Effects were MDSC specific, since WT and CCR2-KO conventional T (Tcon) cells had comparable proliferation and production of inflammatory cytokines, and suppressive functions of WT and CCR2-KO Foxp3+ Treg cells were also similar. We used a thioglycolate-induced peritonitis model to demonstrate a role for CCR2/MCP-1 in trafficking of CCR2+ cells to an inflammatory site, and showed the ability of a CCR2 antagonist to inhibit such trafficking. Use of this CCR2 antagonist promoted anti-tumor immunity and limited tumor growth. In summary, tumor cells are the prime source of MCP-1 that promotes MDSC recruitment, and our genetic and pharmacologic data demonstrate that CCR2 targeting may be an important component of cancer immunotherapy.

Design and Optimization of Sulfone Pyrrolidine Sulfonamide Antagonists of Transient Receptor Potential Vanilloid-4 with in Vivo Activity in a Pulmonary Edema Model.

Pulmonary edema is a common ailment of heart failure patients and has remained an unmet medical need due to dose-limiting side effects associated with current treatments. Preclinical studies in rodents have suggested that inhibition of transient receptor potential vanilloid-4 (TRPV4) cation channels may offer an alternative-and potentially superior-therapy. Efforts directed toward small-molecule antagonists of the TRPV4 receptor have led to the discovery of a novel sulfone pyrrolidine sulfonamide chemotype exemplified by lead compound 6. Design elements toward the optimization of TRPV4 activity, selectivity, and pharmacokinetic properties are described. Activity of leading exemplars 19 and 27 in an in vivo model suggestive of therapeutic potential is highlighted herein.

Synthesis of 5-fluoro- and 5-hydroxymethanoprolines via lithiation of N-BOC-methanopyrrolidines. Constrained Cγ-exo and Cγ-endo Flp and Hyp conformer mimics.

Proline derivatives with a C(γ)-exo pucker typically display a high amide bond trans/cis (K(T/C)) ratio. This pucker enhances n→π* overlap of the amide oxygen and ester carbonyl carbon, which favors a trans amide bond. If there were no difference in n→π* interaction between the ring puckers, then the correlation between ring pucker and K(T/C) might be broken. To explore this possibility, proline conformations were constrained using a methylene bridge. We synthesized discrete gauche and anti 5-fluoro- and 5-hydroxy-N-acetylmethanoproline methyl esters from 3-syn and 3-anti fluoro- and hydroxymethanopyrrolidines using directed α-metalation to introduce the α-ester group. NBO calculations reveal minimal n→π* orbital interactions, so contributions from other forces might be of greater importance in determining K(T/C) for the methanoprolines. Consistent with this hypothesis, greater trans amide preferences were found in CDCl(3) for anti isomers en-MetFlp and en-MetHyp (72-78% trans) than for the syn stereoisomers ex-MetFlp and ex-MetHyp (54-67% trans). These, and other, K(T/C) results that we report here indicate how substituents on proline analogues can affect amide preferences by pathways other than ring puckering and n→π* overlap and suggest that caution should be exercised in assigning enhanced pyrrolidine C(γ)-exo ring puckering based solely on enhanced trans amide preference.

Synthesis of conformationally constrained 5-fluoro- and 5-hydroxymethanopyrrolidines. Ring-puckered mimics of gauche- and anti-3-fluoro- and 3-hydroxypyrrolidines.

N-acetylmethanopyrrolidine methyl ester and its four 5-syn/anti-fluoro and hydroxy derivatives have been synthesized from 2-azabicyclo[2.2.0]hex-5-ene, a 1,2-dihydropyridine photoproduct. These conformationally constrained mimics of idealized C(ß)-gauche and C(ß)-anti conformers of pyrrolidines were prepared in order to determine the inherent bridge bias and subsequent heteroatom substituent effects upon trans/cis amide preferences. The bridgehead position and also the presence of gauche(syn)/anti-5-fluoro or 5-hydroxy substituents have minimal influence upon the K(T/C) values of N-acetylamide conformers in both CDCl(3) (43-54% trans) and D(2)O (53-58% trans). O-Benzoylation enhances the trans amide preferences in CDCl(3) (65% for a syn-OBz, 61% for an anti-OBz) but has minimal effect in D(2)O. The synthetic methods developed for N-BOC-methanopyrrolidines should prove useful in the synthesis of more complex derivatives containing α-ester substituents. The K(T/C) results obtained in this study establish baseline amide preferences that will enable determination of contributions of α-ester substituents to trans-amide preferences in methanoprolines.

Oligomers of a 5-carboxy-methanopyrrolidine ß-amino acid. A search for order.

CD spectra for homooligomers (n = 4, 6, 8) of (1S,4R,5R)-5-syn-carboxy-2-azabicyclo[2.1.1]hexane (MPCA), a methano-bridged pyrrolidine ß-carboxylic acid, suggest an ordered secondary structure. Even in the absence of internal hydrogen bonding, solution NMR, X-ray, and in silico analyses of the tetramer are indicative of conformations with trans-amides and C(5)-amide-carbonyls oriented toward the C(4) bridgehead. This highly constrained ß-amino acid could prove useful in the ongoing development of well-defined foldamers.