Crystal structure and Hirshfeld surface analysis of 3-methyl-4-oxo-N-phenyl-3,4-dihydroquinazoline-2-carbothioamide

The asymmetric unit of the title compound, C16H13N3OS, comprises two molecules (A and B) with similar conformations that differ mainly in the orientation of the phenyl group relative to the rest of the molecule, as expressed by the Cthioamide—Nthioamide—Cphenyl—Cphenyl torsion angle of 49.3 (3)° for molecule A and of 5.4 (3)° for molecule B. In the crystal, two intermolecular N—H...N hydrogen bonds lead to the formation of a dimer with R 2 2(10) graph-set notation. A Hirshfeld surface analysis revealed that H...H interactions are the most important intermolecular interactions, contributing 40.9% to the Hirshfeld surface.

Crystal structure of (+)-(1S,5S,6S,7S,10S,11S,16S)-16-hydroxy-7-(methoxymethoxy)-11,15,18,18-tetramethyl-3,13-dioxo-2,4-dioxatetracyclo[12.3.1.01,5.06,11]octadec-14-en-10-yl benzoate

In the fused tetracyclic system of the title compound, C29H36O9, the five-membered dioxolane ring adopts a twist conformation; the two adjacent C atoms deviate alternately from the mean plane of the other three atoms by −0.252 (6) and 0.340 (6) Å. The cyclohexane, cyclohexene and central cyclooctane rings show chair, half-chair and boat-chair forms, respectively. There are three intramolecular C—H...O interactions supporting the molecular conformation, with one S(6) and two S(7) graph-set motifs. In the crystal, intermolecular O—H...O hydrogen bonds connect the molecules into a helical chain running along the c-axis direction, generating a C(7) graph-set motif. The chains are further linked by intermolecular C—H...O interactions to construct a three-dimensional network. There is no valid C—H...π interaction.

Crystal structure of 9,9-diethyl-9H-fluorene-2,4,7-tricarbaldehyde

The title compound, C20H18O3, crystallizes in the space group P21/c with one molecule in the asymmetric unit of the cell. The fluorene skeleton is nearly planar and the crystal structure is composed of molecular layers extending parallel to the (302) plane. Within a layer, one formyl oxygen atom participates in the formation of a Carene—H...O bond, which is responsible for the formation of an inversion symmetric supramolecular motif of graph set R 2 2(10). A second oxygen atom is involved in an intramolecular Carene—H...O hydrogen bond and is further connected with a formyl hydrogen atom of an adjacent molecule. A Hirshfeld surface analysis indicated that the most important contributions to the overall surface are from H...H (46.9%), O...H (27.9%) and C...H (17.8%) interactions.

Structural and theoretical studies of 4-chloro-2-methyl-6-oxo-3,6-dideuteropyrimidin-1-ium chloride (d 6)

The title compound, C5D6ClN2O+·Cl−, crystallizes in the orthorhombic space group, Pbcm, and consists of a 4-chloro-2-methyl-6-oxo-3,6-dihydropyrimidin-1-ium cation and a chloride anion where both moieties lie on a crystallographic mirror. The cation is disordered and was refined as two equivalent forms with occupancies of 0.750 (4)/0.250 (4), while the chloride anion is triply disordered with occupancies of 0.774 (12), 0.12 (2), and 0.11 (2). Unusually, the bond angles around the C=O unit range from 127.2 (6) to 115.2 (3)° and similar angles have been found in other structures containing a 6-oxo-3,6-dihydropyrimidin-1-ium cation, including the monclinic polymorph of the title compound, which crystallizes in the monoclinic space group P21/c [Kawai et al. (1973). Cryst. Struct. Comm. 2, 663–666]. The cations and anions pack into sheets in the ab plane linked by N—H...Cl hydrogen bonds as well as C—H...O and Cl...O interactions. In graph-set notation, these form R 3 3(11) and R 3 2(9) rings. Theoretical calculations seem to indicate that the reason for the unusual angles at the sp 2 C is the electrostatic interaction between the oxygen atom and the adjacent N—H hydrogen.

Crystal structure of a methanol solvate of a macrocycle bearing two flexible side-arms

Di-tert-butyl N,N′-{[13,15,28,30,31,33-hexaethyl-3,10,18,25,32,34-hexaazapentacyclo[25.3.1.15,8.112,16.120,23]tetratriaconta-1(31),3,5,7,9,12(33),13,15,18,20,22,24,27,29-tetradecaene-14,29-diyl]bis(methylene)}dicarbamate methanol disolvate, C52H72N8O4·2CH3OH, was found to crystallize in the space group P21/c with one half of the macrocycle (host) and one molecule of solvent (guest) in the asymmetric unit of the cell, i.e. the host molecule is located on a crystallographic symmetry center. Within the 1:2 host–guest complex, the solvent molecules are accommodated in the host cavity and held in their positions by O—H...N and N—H...O bonds, thus forming ring synthons of graph set R 2 2(7). The connection of the 1:2 host-guest complexes is accomplished by C—H...O, C—H...N and C—H...π interactions, which create a three-dimensional supramolecular network.

Tetraaquabis(pyridine-3-carbonitrile-κN 1)nickel(II) benzene-1,4-dicarboxylate tetrahydrate

A nickel(II) terephthalate complex, viz. [Ni(C6H4N2)2(H2O)4](O2CC6H4CO2)·4H2O, has been synthesized and studied by single-crystal X-ray diffraction. It crystallizes in the triclinic space group P\overline{1}. The crystal structure shows an approximately octahedral coordination environment of the complex with the [Ni(H2O)4(3-NCpy)2]2+ (3-NCpy is pyridine-3-carbonitrile) cation associated with four free water molecules and hydrogen bonded to a terephthalate dianion [graph set R 2 2(8)]. The supramolecular structure of the compound is stabilized by a three-dimensional array of O—H...O and O—H...N hydrogen bonds, along with π–π stacked pyridine-3-carbonitrile rings and C—H...O interactions.

Network-principled deep generative models for designing drug combinations as graph sets

Motivation Combination therapy has shown to improve therapeutic efficacy while reducing side effects. Importantly, it has become an indispensable strategy to overcome resistance in antibiotics, antimicrobials and anticancer drugs. Facing enormous chemical space and unclear design principles for small-molecule combinations, computational drug-combination design has not seen generative models to meet its potential to accelerate resistance-overcoming drug combination discovery. Results We have developed the first deep generative model for drug combination design, by jointly embedding graph-structured domain knowledge and iteratively training a reinforcement learning-based chemical graph-set designer. First, we have developed hierarchical variational graph auto-encoders trained end-to-end to jointly embed gene–gene, gene–disease and disease–disease networks. Novel attentional pooling is introduced here for learning disease representations from associated genes’ representations. Second, targeting diseases in learned representations, we have recast the drug-combination design problem as graph-set generation and developed a deep learning-based model with novel rewards. Specifically, besides chemical validity rewards, we have introduced novel generative adversarial award, being generalized sliced Wasserstein, for chemically diverse molecules with distributions similar to known drugs. We have also designed a network principle-based reward for disease-specific drug combinations. Numerical results indicate that, compared to state-of-the-art graph embedding methods, hierarchical variational graph auto-encoder learns more informative and generalizable disease representations. Results also show that the deep generative models generate drug combinations following the principle across diseases. Case studies on four diseases show that network-principled drug combinations tend to have low toxicity. The generated drug combinations collectively cover the disease module similar to FDA-approved drug combinations and could potentially suggest novel systems pharmacology strategies. Our method allows for examining and following network-based principle or hypothesis to efficiently generate disease-specific drug combinations in a vast chemical combinatorial space. Availability and implementation https://github.com/Shen-Lab/Drug-Combo-Generator. Supplementary information Supplementary data are available at Bioinformatics online.

Synthesis and crystal structures of 2-(ferrocenylcarbonyl)benzoic acid and 3-ferrocenylphthalide

The title compounds, 2-(ferrocenylcarbonyl)benzoic acid, [Fe(C5H5)(C13H9O3)], 1, and 3-ferrocenylphthalide [systematic name: 3-ferrocenyl-2-benzofuran-1(3H)-one], [Fe(C5H5)(C13H9O2)], 2, have been synthesized and structurally characterized by single-crystal X-ray diffraction. The crystal structure of compound 1 was solved recently at room temperature [Qin, Y. (2019). CSD Communication (CCDC deposition number 1912662). CCDC, Cambridge, England]. Here we report a redetermination of its crystal structure at 90 K with improved precision by a factor of about three. The molecular structures of both compounds exhibit a typical sandwich structure. In the crystal packing of compound 1, each molecule engages in intermolecular hydrogen bonding, forming a centrosymmetric dimer with graph-set notation R 2 2 (8) and an O...O distance of 2.6073 (15) Å. There are weak C—H...O and C—H...π interactions in the crystal packing of compound 2. The phthalide moiety in 2 is oriented roughly perpendicular to the ferrocene backbone, with a dihedral angle of 77.4 (2)°.

Network-principled deep generative models for designing drug combinations as graph sets

MotivationCombination therapy has shown to improve therapeutic efficacy while reducing side effects. Importantly, it has become an indispensable strategy to overcome resistance in antibiotics, anti-microbials, and anti-cancer drugs. Facing enormous chemical space and unclear design principles for small-molecule combinations, computational drug-combination design has not seen generative models to meet its potential to accelerate resistance-overcoming drug combination discovery.ResultsWe have developed the first deep generative model for drug combination design, by jointly embedding graph-structured domain knowledge and iteratively training a reinforcement learning-based chemical graph-set designer. First, we have developed Hierarchical Variational Graph Auto-Encoders (HVGAE) trained end-to-end to jointly embed gene-gene, gene-disease, and disease-disease networks. Novel attentional pooling is introduced here for learning disease-representations from associated genes’ representations. Second, targeting diseases in learned representations, we have recast the drug-combination design problem as graph-set generation and developed a deep learning-based model with novel rewards. Specifically, besides chemical validity rewards, we have introduced novel generative adversarial award, being generalized sliced Wasserstein, for chemically diverse molecules with distributions similar to known drugs. We have also designed a network principle-based reward for drug combinations. Numerical results indicate that, compared to state-of-the-art graph embedding methods, HVGAE learns more informative and generalizable disease representations. Results also show that the deep generative models generate drug combinations following the principle across diseases. Case studies on four diseases show that network-principled drug combinations tend to have low toxicity. The generated drug combinations collectively cover the disease module similar to FDA-approved drug combinations and could potentially suggest novel systems-pharmacology strategies. Our method allows for examining and following network-based principle or hypothesis to efficiently generate disease-specific drug combinations in a vast chemical combinatorial space.Availabilityhttps://github.com/Shen-Lab/Drug-Combo-Generator