Search

Structure-based virtual screening (SBVS) is a key workflow in computational drug discovery. SBVS models are assessed by measuring the enrichment of known active molecules over decoys in retrospective screens. However, the standard formula for enrichment cannot estimate model performance on very large libraries. Additionally, current screening benchmarks cannot easily be used with machine learning (ML) models due to data leakage. We propose an improved formula for calculating VS enrichment and introduce the BayesBind benchmarking set composed of protein targets that are structurally dissimilar to those in the BigBind training set. We assess current models on this benchmark and find that none perform appreciably better than a KNN baseline., Comment: 10 pages, 4 figures, and 4 tables. The source code is available at https://github.com/molecularmodelinglab/bigbind

Nearest neighbor-based similarity searching is a common task in chemistry, with notable use cases in drug discovery. Yet, some of the most commonly used approaches for this task still leverage a brute-force approach. In practice this can be computationally costly and overly time-consuming, due in part to the sheer size of modern chemical databases. Previous computational advancements for this task have generally relied on improvements to hardware or dataset-specific tricks that lack generalizability. Approaches that leverage lower-complexity searching algorithms remain relatively underexplored. However, many of these algorithms are approximate solutions and/or struggle with typical high-dimensional chemical embeddings. Here we evaluate whether a combination of low-dimensional chemical embeddings and a k-d tree data structure can achieve fast nearest neighbor queries while maintaining performance on standard chemical similarity search benchmarks. We examine different dimensionality reductions of standard chemical embeddings as well as a learned, structurally-aware embedding -- SmallSA -- for this task. With this framework, searches on over one billion chemicals execute in less than a second on a single CPU core, five orders of magnitude faster than the brute-force approach. We also demonstrate that SmallSA achieves competitive performance on chemical similarity benchmarks.

Modeling the properties of chemical mixtures is a difficult but important part of any modeling process intended to be applicable to the often messy and impure phenomena of everyday life, including food and environmental safety, healthcare, etc. Part of this difficulty stems from the increased complexity of designing suitable model validation schemes for mixture data, a fact which has been elucidated in previous work only in the case of binary mixture models. We extend these previously defined validation strategies for QSAR modeling of binary mixtures to the more complex case of general, $N$-ary mixtures and argue that these strategies are applicable to many modeling tasks beyond simple chemical mixtures. Additionally, we propose a method of establishing a baseline model performance for each mixture dataset to be in used in model selection comparisons. This baseline is intended to account for the statistical dependence generically present between the properties of mixtures that share constituents. We contend that without such a baseline, estimates of model performance can be dramatically overestimated, and we demonstrate this with multiple case studies using real and simulated data., Comment: 22 pages, 1 figure

Molecular docking aims to predict the 3D pose of a small molecule in a protein binding site. Traditional docking methods predict ligand poses by minimizing a physics-inspired scoring function. Recently, a diffusion model has been proposed that iteratively refines a ligand pose. We combine these two approaches by training a pose scoring function in a diffusion-inspired manner. In our method, PLANTAIN, a neural network is used to develop a very fast pose scoring function. We parameterize a simple scoring function on the fly and use L-BFGS minimization to optimize an initially random ligand pose. Using rigorous benchmarking practices, we demonstrate that our method achieves state-of-the-art performance while running ten times faster than the next-best method. We release PLANTAIN publicly and hope that it improves the utility of virtual screening workflows., Comment: Camera-ready submission to ICML CompBio workshop. 5 pages and 1 figure

The growing families of two-dimensional crystals derived from naturally occurring van der Waals materials offer an unprecedented platform to investigate elusive physical phenomena and could be of use in a diverse range of devices. Of particular interest are recently reported atomic sheets of non-van der Waals materials, which could allow a better comprehension of the nature of structural bonds and increase the functionality of prospective heterostructures. Here, we study the optostructural properties of ultrathin non-van der Waals InGaS3 sheets produced by standard mechanical cleavage. Our ab initio calculation results suggest an emergence of authentically delicate out-of-plane covalent bonds within its unit cell, and, as a consequence, an artificial generation of layered structure within the material. Those yield to singular layer isolation energies of around 50 meVA-2, which is comparable with the conventional van der Waals material's monolayer isolation energies of 20 - 60 meVA-2. In addition, we provide a comprehensive analysis of the structural, vibrational, and optical properties of the materials presenting that it is a wide bandgap (2.73 eV) semiconductor with a high-refractive index (higher than 2.5) and negligible losses in the visible and infrared spectral ranges. It makes it a perfect candidate for further establishment of visible-range all-dielectric nanophotonics.

A boost in the development of flexible and wearable electronics facilitates the design of new materials to be applied as transparent conducting films (TCFs). Although single-walled carbon nanotube (SWCNT) films are the most promising candidates for flexible TCFs, they still do not meet optoelectronic requirements demanded their successful industrial integration. In this study, we proposed and thoroughly investigated a new approach that comprises simultaneous bilateral (outer and inner surfaces) SWCNT doping after their opening by thermal treatment at 400 C under an ambient air atmosphere. Doping by a chloroauric acid (HAuCl$_{4}$) ethanol solution allowed us to achieve the record value of sheet resistance of 31 $\pm$ 4 $\Omega$/sq at a transmittance of 90% in the middle of visible spectra (550 nm). The strong p-doping was examined by open-circuit potential (OCP) measurements and confirmed by ab initio calculations demonstrating a downshift of Fermi level around 1 eV for the case of bilateral doping.

Violation of time reversal and spatial inversion symmetries has profound consequences for elementary particles and cosmology. Spontaneous breaking of these symmetries at phase transitions gives rise to unconventional physical phenomena in condensed matter systems, such as ferroelectricity induced by magnetic spirals, electromagnons, non-reciprocal propagation of light and spin waves, and the linear magnetoelectric (ME) effect - the electric polarization proportional to the applied magnetic field and the magnetization induced by the electric field. Here, we report the experimental study of the holmium-doped langasite, Ho$_{x}$La$_{3-x}$Ga$_5$SiO$_{14}$, showing a puzzling combination of linear and highly non-linear ME responses in the disordered paramagnetic state: its electric polarization grows linearly with the magnetic field but oscillates many times upon rotation of the magnetic field vector. We propose a simple phenomenological Hamiltonian describing this unusual behavior and derive it microscopically using the coupling of magnetic multipoles of the rare-earth ions to the electric field., Comment: 8 pages, 3 figures

The chemical inertness of the defect-free basal plane confers environmental stability to MoS2 single-layers, but it also limits their chemical versatility and catalytic activity. The stability of the pristine MoS2 basal plane against oxidation under ambient conditions is a widely accepted assumption in the interpretation of various studies and applications. However, single-atom level structural investigations reported here reveal that oxygen atoms spontaneously incorporate into the basal plane of MoS2 single layers during ambient exposure. Our scanning tunneling microscopy investigations reveal a slow oxygen substitution reaction, upon which individual sulfur atoms are one by one replaced by oxygen, giving rise to solid solution type 2D MoS2-xOx crystals. O substitution sites present all over the basal plane act as single-atomic active reaction centers, substantially increasing the catalytic activity of the entire MoS2 basal plane for the electrochemical H2 evolution reaction., Comment: 6 pages, 5 figures