Atom Perturber Research Articles

Interpretable Deep-Learning pKa Prediction for Small Molecule Drugs via Atomic Sensitivity Analysis.

Machine learning (ML) models now play a crucial role in predicting properties essential to drug development, such as a drug's logscale acid-dissociation constant (pKa). Despite recent architectural advances, these models often generalize poorly to novel compounds due to a scarcity of ground-truth data. Further, these models lack interpretability. To this end, with deliberate molecular embeddings, atomic-resolution information is accessible in chemical structures by observing the model response to atomic perturbations of an input molecule. Here, we present BCL-XpKa, a deep neural network (DNN)-based multitask classifier for pKa prediction that encodes local atomic environments through Mol2D descriptors. BCL-XpKa outputs a discrete distribution for each molecule, which stores the pKa prediction and the model's uncertainty for that molecule. BCL-XpKa generalizes well to novel small molecules. BCL-XpKa performs competitively with modern ML pKa predictors, outperforms several models in generalization tasks, and accurately models the effects of common molecular modifications on a molecule's ionizability. We then leverage BCL-XpKa's granular descriptor set and distribution-centered output through atomic sensitivity analysis (ASA), which decomposes a molecule's predicted pKa value into its respective atomic contributions without model retraining. ASA reveals that BCL-XpKa has implicitly learned high-resolution information about molecular substructures. We further demonstrate ASA's utility in structure preparation for protein-ligand docking by identifying ionization sites in 93.2% and 87.8% of complex small molecule acids and bases. We then applied ASA with BCL-XpKa to identify and optimize the physicochemical liabilities of a recently published KRAS-degrading PROTAC.

Excitation Dependent Multiemissive Behaviour of Triimidazole Carboxylic Acid anti‐Kasha Luminophore and its Mononuclear Metal Complexes

AbstractMaterials based on small molecules characterized by multi‐component short and long‐lived emissive behaviour are extremely desirable for various applications. The derivative of cyclic triimidazole (TT) functionalized with a carboxylic group, namely TT‐COOH, is isolated, investigated and used as coordinative ligand to obtain a series of isomorphous mononuclear complexes of general formula [M(TT‐COO)2(H2O)2] (M=Mn(II), Co(II), Ni(II), Cu(II), Zn(II) and Cd(II)). Crystals of TT‐COOH display aggregation induced enhanced emission (AIEE) comprising dual fluorescence and dual phosphorescence of molecular origin together with an aggregation induced long lived, low energy phosphorescence. Emissive features are retained in crystals of the Zn(II) and Cd(II) complexes, where the metal plays the role of an external heavy atom perturber, but strongly quenched in the others. The mechanisms involved in the photophysics of TT‐COOH and its metal complexes have been disclosed thanks to combined structural, spectroscopic and computational investigations which reveal the presence of anti‐Kasha emissions.

VARPA: In Silico Additive Screening for Protein-Based Lighting Devices.

Protein optoelectronics is an emerging field facing implementation and stabilization challenges of proteins in harsh non-natural environments, such as dry polymers, inorganic materials, etc., operating at high temperatures/irradiations. In this context, additives promoting structural and functional protein stabilization are paramount to realize highly performing devices. On one hand, trial-error experimental assays based on previous knowledge of classical additives in aqueous solutions are effort/time-consuming, while their translation to water-less matrices is uncertain. On the other hand, computational simulations (molecular dynamics, electronic structure methods, etc.) are limited by the system size and time. Herein, ligand-binding affinity and atomic perturbations to create a day-fast computational method combining Vina And Rosetta for Protein Additives (VARPA) to simulate the stabilization effect of sugars for the archetypal enhanced green fluorescent protein embedded in a standard dry polymer color-converting filter for bio-hybrid light-emitting diodes is merged. The VARPA's sugar additive prediction trend for protein stabilization is nicely validated by thermal and photophysical studies as well as lighting device analysis. The device stability followed the predicted enhanced stability trend, reaching a 40-fold improvement compared to reference devices. Overall, VARPA can be adapted to a myriad of additives and proteins, driving first-step experimental efforts toward highly performing protein devices.

Topology-aware universal adversarial attack on 3D object tracking

3D object tracking based on deep neural networks has a wide range of potential applications, such as autonomous driving and robotics. However, deep neural networks are vulnerable to adversarial examples. Traditionally, adversarial examples are generated by applying perturbations to individual samples, which requires exhaustive calculations for each sample and thereby suffers from low efficiency during malicious attacks. Hence, the universal adversarial perturbation has been introduced, which is sample-agnostic. The universal perturbation is able to make classifiers misclassify most samples. In this paper, a topology-aware universal adversarial attack method against 3D object tracking is proposed, which can lead to predictions of a 3D tracker deviating from the ground truth in most scenarios. Specifically, a novel objective function consisting of a confidence loss, direction loss and distance loss generates an atomic perturbation from a tracking template, and aims to fail a tracking task. Subsequently, a series of atomic perturbations are iteratively aggregated to derive the universal adversarial perturbation. Furthermore, in order to address the characteristic of permutation invariance inherent in the point cloud data, the topology information of the tracking template is employed to guide the generation of the universal perturbation, which imposes correspondences between consecutively generated perturbations. The generated universal perturbation is designed to be aware of the topology of the targeted tracking template during its construction and application, thus leading to superior attack performance. Experiments on the KITTI dataset demonstrate that the performance of 3D object tracking can be significantly degraded by the proposed method.

Electronic and vibrational spectroscopies of aromatic clusters with He in a supersonic jet: The case of neutral and cationic phenol-Hen (n = 1 and 2).

Van der Waals clusters composed of He and aromatic molecules provide fundamental information about intermolecular interactions in weakly bound systems. In this study, phenol-helium clusters (PhOH-Hen with n ≤ 2) are characterized for the first time by UV and IR spectroscopies. The S1 ← S0 origin and ionization energy both show small but additive shifts, suggesting π-bound structures of these clusters, a conclusion supported by rotational contour analyses of the S1 origin bands. The OH stretching vibrations of the PhOH moiety in the clusters match with those of bare PhOH in both the S0 and D0 states, illustrating the negligible perturbation of the He atoms on the molecular vibration. Matrix shifts induced by He attachment are discussed based on the observed band positions with the help of complementary quantum chemical calculations. For comparison, the UV and ionization spectra of PhOH-Ne are reported as well.

Size and strain modulation of dielectric constant on atomic bond relaxation

From the viewpoints of bond order-length-strength correlation, core–shell structural model, and local bond average approach, we examined the size and strain effects on the dielectric constant of the transition metal dichalcogenides system. Consistency in theoretical results and reported values confirms that: (i) the surface atomic coordination number deficiency and bond energy perturbation dictate the size effect of the dielectric constant for nanometric semiconductors, and (ii) the bond elongation and softening lead to the tensile strain-induced rise in dielectric constant. The analytical function of dielectric constant dependence on size and strain is beyond the scope of available approaches, which not only provides a new understanding of the physical mechanism of the dielectric response to perturbations but also is helpful in the quantitative design of optoelectronic and photovoltaic nanodevices.

Analysis of coordinated NMR chemical shifts to map allosteric regulatory networks in proteins.

The exquisite sensitivity of the NMR chemical shift to local environment makes it an ideal probe to assess atomic level perturbations in proteins of all sizes and structural compositions. Recent advances in solution and solid-state NMR spectroscopy of biomolecules have leveraged the chemical shift to report on short- and long-range couplings between individual amino acids to establish "networks" of residues that form the basis of allosteric pathways that transmit chemical signals through the protein matrix to induce functional responses. The simple premise that thermodynamically and functionally coupled regions of a protein (i.e. active and allosteric sites) should be reciprocally sensitive to structural or dynamic perturbations has enabled NMR spectroscopy, the premier method for molecular resolution of protein structural fluctuations, to occupy a place at the forefront of investigations into protein allostery. Here, we detail several key methods of NMR chemical shift analysis to extract mechanistic information about long-range chemical signaling in a protein, focusing on practical methodological aspects and the circumstances under which a given approach would be relevant. We also detail some of the experimental considerations that should be made when applying these methods to specific protein systems.

The Features of Infrared Absorption of Boron‐Doped Silicon

Herein, the influence of heat treatment at 400 °C on the spectrum of boron intracenter transitions in silicon using IR absorption spectroscopy is investigated. In the transition region from the ground 1Γ8+ state associated with the p3/2 valence band of Si to the odd‐parity excited states of boron, a new absorption line with its maximum at 261.3 cm−1 is observed in the thermally treated boron‐doped Cz‐Si. Oxygen is a component of defect that is responsible for the detected absorption line. Perturbation of boron atoms due to the inhomogeneous stress effect from neighboring oxygen atoms results in a frequency shift in the main boron transition. The defect associated with 261.3 cm−1 line is also observed in as‐grown silicon. The defect disappears during annealing at 550 °C. The concentration of defects and binding energy of the 1Γ8+ ground state are estimated assuming that the line observed belongs to the transition of boron 1Γ8+ → 2Γ8−, subjected to the perturbation from a neighboring oxygen atom.

In-plane anisotropic electronic properties in layered α′-In2Se3

In2Se3 polymorphs have been extensively studied because of their diverse physical properties such as piezoelectricity, photoelectricity, and ferroelectricity, thereby showing plentiful promising applications in integrated electronic devices. These diverse properties are strongly dependent on or affected by their atomic bonding arrangement in the crystal phases. Combining lattice symmetry and local atomic perturbation, we demonstrate a novel layered α′-In2Se3 phase by using the first-principles calculations, which is reconstructed from the inverted tetrahedral bonding configuration by the in-plane displacive middle layer Se atom. The optimized structure of monolayer α′-In2Se3 has triple degenerated atomic configurations with different Se atom orientations. We noted that these degenerated atomic configurations exhibit a moderate switching barrier (about 61 meV/f.u.) between them. To further explore this atom-oriented anisotropic property in α′-In2Se3, the electronic properties were studied with an orthorhombic unit cell. The comparative results for the orthogonal Se atom orientations suggest that the nonbonding orbital coupling of the displacive Se atoms induces large in-plane anisotropic optical absorption and electrical transport properties. This study of the layered α′-In2Se3 phase can extend the realm of switchable anisotropic optoelectronic applications in future electronic devices.

Long-range Rydberg molecule Rb2 : Two-electron R -matrix calculations at intermediate internuclear distances

The adiabatic potential energy curves of Rb$_2$ in the long-range Rydberg electronic states are calculated using the two-electron \textit{R}-matrix method [M. Tarana, R. \v{C}ur\'{i}k, Phys. Rev. A \textbf{93}, 012515 (2016)] for the intermediate internuclear separations between 35 a.u. and 200 a.u. The results are compared with the zero-range models to find a region of the internuclear distances where the Fermi's pseudopotential approach provides accurate energies. A finite-range potential model of the atomic perturber is used to calculate the wave functions of the Rydberg electron and their features specific for the studied range of internuclear distances are identified.

Efficient Analytic Second Derivative of Electrostatic Embedding QM/MM Energy: Normal Mode Analysis of Plant Cryptochrome.

Analytic second derivatives of electrostatic embedding (EE) quantum mechanics/molecular mechanics (QM/MM) energy are important for performing vibrational analysis and simulating vibrational spectra of quantum systems interacting with an environment represented as a classical electrostatic potential. The main bottleneck of EE-QM/MM second derivatives is the solution of coupled perturbed equations for each MM atom perturbation. Here, we exploit the Q-vector method [J. Chem. Phys., 2019, 151, 041102] to workaround this bottleneck. We derive the full analytic second derivative of the EE-QM/MM energy, which allows us to compute QM, MM, and QM-MM Hessian blocks in an efficient and easy to implement manner. To show the capabilities of our method, we compute the normal modes for the full Arabidopsis thaliana plant cryptochrome. We show that the flavin adenine dinucleotide vibrations (QM subsystem) strongly mix with protein modes. We compute approximate vibronic couplings for the lowest bright transition, from which we extract spectral densities and the homogeneous broadening of FAD absorption spectrum in protein using vibrationally resolved electronic spectrum simulations.

Perturbations of Christoffel–Darboux Kernels: Detection of Outliers

Two central objects in constructive approximation, the Christoffel–Darboux kernel and the Christoffel function, encode ample information about the associated moment data and ultimately about the possible generating measures. We develop a multivariate theory of the Christoffel–Darboux kernel in $$\mathbb {C}^d$$ , with emphasis on the perturbation of Christoffel functions and their level sets with respect to perturbations of small norm or low rank. The statistical notion of leverage score provides a quantitative criterion for the detection of outliers in large data. Using the refined theory of Bergman orthogonal polynomials, we illustrate the main results, including some numerical simulations, in the case of finite atomic perturbations of area measure of a 2D region. Methods of function theory of a complex variable and (pluri) potential theory are widely used in the derivation of our perturbation formulas.

Bent Allenes or Di-1,3-betaines-An Answer Given on the Magnetic Criterion.

The spatial magnetic properties, through-space NMR shieldings (TSNMRS), of bent allene 1, the corresponding C-extended 1,3-butadiene derivative 2, and a number of related compounds 3-20 have been calculated using the gauge-independent atomic orbital perturbation method, employing the nucleus-independent chemical shift concept and visualized as isochemical shielding surfaces of various sizes and directions. Prior to that, both structures and 13C chemical shifts were calculated and compared with available experimental bond lengths and δ(13C)/ppm values (also, as a quality criterion for the computed structures). Bond lengths, the δ(13C)/ppm, and the TSNMRS values are employed to qualify and quantify the electronic structure of the studied compounds in terms of dative or classical electron-sharing bonds.

Voronoi‐Based Similarity Distances between Arbitrary Crystal Lattices

AbstractThis paper develops a new continuous approach to a similarity between periodic lattices of ideal crystals. Quantifying a similarity between crystal structures is needed to substantially speed up the crystal structure prediction, because the prediction of many target properties of crystal structures is computationally slow and is essentially repeated for many nearly identical simulated structures. The proposed distances between arbitrary periodic lattices of crystal structures are invariant under all rigid motions, satisfy the metric axioms and continuity under atomic perturbations. The above properties make these distances ideal tools for clustering and visualizing large datasets of crystal structures. All the conclusions are rigorously proved and justified by experiments on real and simulated crystal structures.

Suppression of Iodide Ion Migration via Sb2S3 Interfacial Modification for Stable Inorganic Perovskite Solar Cells.

In mixed halide perovskite, the halide phase segregation is commonly observed due to halide ion migration, which causes severe stability issues in perovskite devices. Here, we directly revealed the iodide-migration process via potentiostatic treatment in CsPbIBr2 perovskite. The absence of iodide ions was reduced significantly via a Sb2S3 interfacial modification. We further employed the density functional theory (DFT) calculation to optimize the geometry positions at the perovskite interface and radial distribution functions (RDF) to analyze the atom perturbation. The simulation yielded a slight distortion of the perovskite lattice at the Sb2S3-CsPbIBr2 interface and iodide ion fluctuation was reduced due to the decrease of halide vacancies. In addition, the thermally stimulated current was calculated to evaluate the defect density in the modified perovskite device. Due to the Sb2S3 interaction with perovskite, the device became stable against humidity and maintained its photoactivity over 400 h. The champion efficiency of 9.31% with 26.31% improvement was obtained in modified CsPbIBr2 perovskite solar cells.

Heavy atom perturbation by the incorporation of iodine ion into BaTiO3 lattice: Reduction of fluorescence and enhancement of rate of interfacial charge transfer process under the visible light irradiation

Iodine was incorporated into the BaTiO3 (BTO) lattice in the concentration range of 0.3-0.9 wt. % and the photocatalytic activities were tested for the degradation of congo red (CR) as a model compound under visible light irradiation. The doped iodine (I5+) plays a major role in improving the efficiency of the charge transfer dynamics of the system. I5+ is substituted at Ti4+ site in BTO lattice and it extends the photoresponse of the catalyst to the visible light range. 0.7 wt. % of I doped photocatalyst shows high photocatalytic activity due to the synergistic effects that include optimum dopant concentration, enhanced visible light absorption, efficient charge carrier separation. I induced dopant energy level facilitates the absorption of visible light and the excitation of electrons takes place from the valence band (VB) to the dopant energy level. I being a heavy atom can perturb the fluorescence process and in turn increases the lifetime of the photogenerated charge carriers. The active role of hole, hydroxyl radical and superoxide radicals in the degradation reaction was determined by the addition of scavengers like methanol (hydroxyl radical scavenger), benzoquinone (superoxide radical scavenger) and oxalic acid (hole scavenger).

Room-Temperature Phosphorescence from Encapsulated Pyrene Induced by Xenon.

Phosphorescence from pyrene especially at room temperature is uncommon. This emission was recorded utilizing a supramolecular organic host and the effect due to the heavy atom. Poor intersystem crossing from S1 to T1, small radiative rate constant from T1, and large rate constant for oxygen quenching hinder the phosphorescence of aromatic molecules at room temperature in solution. In this study, these limitations are overcome by encapsulating a pyrene molecule within a water-soluble capsule (octa acid, OA) and purging with xenon. While OA suppressed oxygen quenching, xenon enabled the intersystem crossing from S1 to T1 and radiative process from T1 to S0 through the well-known heavy atom effect. The close interaction facilitated between the pyrene and the heavy atom perturber xenon in the three-component supramolecular assembly (OA, pyrene, and xenon) resulted in phosphorescence from pyrene. Computational modeling and NMR studies supported the postulate that pyrene and more than one molecule of xenon are present within a confined space of the OA capsule.

Influence of temperature on the displacement threshold energy in graphene

The atomic structure of nanomaterials is often studied using transmission electron microscopy. In addition to image formation, the energetic electrons impinging on the sample may also cause damage. In a good conductor such as graphene, the damage is limited to the knock-on process caused by elastic electron-nucleus scattering. This process is determined by the kinetic energy an atom needs to be sputtered, i.e. its displacement threshold energy Ed. This is typically assumed to have a fixed value for all electron impacts on equivalent atoms within a crystal. Here we show using density functional tight-binding simulations that the displacement threshold energy is affected by thermal perturbations of atoms from their equilibrium positions. This effect can be accounted for in the estimation of the displacement cross section by replacing the constant threshold energy value with a distribution. Our refined model better describes previous precision measurements of graphene knock-on damage, and should be considered also for other low-dimensional materials.

Density functional perturbation theory within noncollinear magnetism

We extend the density functional perturbation theory formalism to the case of non-collinear magnetism. The main problem comes with the exchange-correlation (XC) potential derivatives, which are the only ones that are affected by the non-collinearity of the system. Most of the present XC functionals are constructed at the collinear level, such that the off-diagonal (containing magnetization densities along $x$ and $y$ directions) derivatives cannot be calculated simply in the non-collinear framework. To solve this problem, we consider here possibilities to transform the non-collinear XC derivatives to a local collinear basis, where the $z$ axis is aligned with the local magnetization at each point. The two methods we explore are i) expanding the spin rotation matrix as a Taylor series, ii) evaluating explicitly the XC for the local density approximation through an analytical expression of the expansion terms. We compare the two methods and describe their practical implementation. We show their application for atomic displacement and electric field perturbations at the second order, within the norm-conserving pseudopotential methods.

Benzenium Ion: Aromatic as the π-Complex or Antiaromatic as the σ-Complex Being Somewhat Similar to the Cyclopentadienyl Cation.

The spatial magnetic properties, through-space NMR shieldings (TSNMRSs), of the benzenium cation (C6H7+) 1 and of ±I/M-substituted analogues C6H6X+ 3-8 [X = -Me, -CF3, -NH2, -NO2, -NO, -SiH3] have been calculated using the gauge-independent atomic orbital perturbation method employing the nucleus-independent chemical shift concept, and iso-chemical-shielding surfaces of various sizes and directions have been observed. The TSNMRS values were employed to compare the spatial magnetic properties (TSNMRS) of benzene and the benzenium ion 1 and then further compared with analogues 3-8, to answer the question whether the electronic structures of 1 and 3-8 are still similar to those of aromatic species or somewhat similar to the antiaromatic cyclopentadienyl cation 2, supported by structural data and δ(13C)/ppm values.