MolE8: finding DFT potential energy surface minima values from force-field optimised organic molecules with new machine learning representations

The use of machine learning techniques in computational chemistry has gained significant momentum since large molecular databases are now readily available. Predictions of molecular properties using machine learning have advantages over the traditional quantum mechanics calculations because they can be cheaper computationally without losing the accuracy. We present a new extrapolatable and explainable molecular representation based on bonds, angles and dihedrals that can be used to train machine learning models. The trained models can accurately predict the electronic energy and the free energy of small organic molecules with atom types C, H N and O, with a mean absolute error of 1.2 kcal mol⁻¹. The models can be extrapolated to larger organic molecules with an average error of less than 3.7 kcal mol⁻¹ for 10 or fewer heavy atoms, which represent a chemical space two orders of magnitude larger. The rapid energy predictions of multiple molecules, up to 7 times faster than previous ML models of similar accuracy, has been achieved by sampling geometries around the potential energy surface minima. Therefore, the input geometries do not have to be located precisely on the minima and we show that accurate density functional theory energy predictions can be made from force-field optimised geometries with a mean absolute error 2.5 kcal mol⁻¹.

New representations and machine learning calculate DFT minima from force field geometries.     

Improvement of Biological Effects of Root-Filling Materials for Primary Teeth by Incorporating Sodium Iodide

Therapeutic iodoform (CHI₃) is commonly used as a root-filling material for primary teeth; however, the side effects of iodoform-containing materials, including early root resorption, have been reported. To overcome this problem, a water-soluble iodide (NaI)-incorporated root-filling material was developed. Calcium hydroxide, silicone oil, and NaI were incorporated in different weight proportions (30:30:X), and the resulting material was denoted DX (D5~D30), indicating the NaI content. As a control, iodoform instead of NaI was incorporated at a ratio of 30:30:30, and the material was denoted I30. The physicochemical (flow, film thickness, radiopacity, viscosity, water absorption, solubility, and ion releases) and biological (cytotoxicity, TRAP, ARS, and analysis of osteoclastic markers) properties were determined. The amount of iodine, sodium, and calcium ion releases and the pH were higher in D30 than I30, and the highest level of unknown extracted molecules was detected in I30. In the cell viability test, all groups except 100% D30 showed no cytotoxicity. In the 50% nontoxic extract, D30 showed decreased osteoclast formation compared with I30. In summary, NaI-incorporated materials showed adequate physicochemical properties and low osteoclast formation compared to their iodoform-counterpart. Thus, NaI-incorporated materials may be used as a substitute for iodoform-counterparts in root-filling materials after further (pre)clinical investigation.     

Phosphate anion chelation and base-pairing. Design of receptors and carriers for nucleotides and nucleotide analogues

Abstract

A rational approach to the development of effective carriers for the through-membrane transport of GMP (guanosine 5′-monophosphate) at neutral pH is described. The approach detailed is a ditopic one predicated on the use of nucleic acid-base ‘nucleobase’ subunits to provide stabilizing hydrogen bonding interactions and the use of expanded porphyrin anion binding subunits to provide phosphate chelation. Appropriate background studies along with the synthesis of a functioning state-of-the-art system are reported. In addition, the reasons for preparing such GMP transport systems are presented in full.     

Biomimetic Synthesis of Materials for Technology

In a world with ever decreasing natural reserves, researchers are striving to find sustainable methods of producing components for technology. Bioinspired, biokleptic and biomimetic materials can be used to form a wide range of technologically relevant materials under environmentally friendly conditions. Here we investigate a range of biotemplated and bioinspired materials that can be used to develop components for devices, such as optics, photonics, photovoltaics, circuits and data storage.     

High dynamic range proteome imaging with the structured illumination gel imager

A current challenge for proteomics is detecting proteins over the large concentration ranges found in complex biological samples such as whole‐cell extracts. Currently, no unbiased, whole‐proteome analysis scheme is capable of detecting the full range of cellular proteins. This is due in part to the limited dynamic range of the detectors used to sense proteins or peptides. We present a new technology, structured illumination (SI) gel imager, which detects fluorescently labeled proteins in electrophoretic gels over a 1 000 000‐fold concentration range. SI uses computer‐generated masks to attenuate the illumination of highly abundant proteins, allowing for long exposures of low‐abundance proteins, thus avoiding detector saturation. A series of progressively masked gel images are assembled into a single, very high dynamic range image. We demonstrate that the SI imager can detect proteins over a concentration range of approximately 1 000 000‐fold, making it a useful tool for comprehensive, unbiased proteome‐wide surveys.