Generating 3D molecules conditional on receptor binding sites with deep generative models.

The goal of structure-based drug discovery is to find small molecules that bind to a given target protein. Deep learning has been used to generate drug-like molecules with certain cheminformatic properties, but has not yet been applied to generating 3D molecules predicted to bind to proteins by sampling the conditional distribution of protein-ligand binding interactions. In this work, we describe for the first time a deep learning system for generating 3D molecular structures conditioned on a receptor binding site. We approach the problem using a conditional variational autoencoder trained on an atomic density grid representation of cross-docked protein-ligand structures. We apply atom fitting and bond inference procedures to construct valid molecular conformations from generated atomic densities. We evaluate the properties of the generated molecules and demonstrate that they change significantly when conditioned on mutated receptors. We also explore the latent space learned by our generative model using sampling and interpolation techniques. This work opens the door for end-to-end prediction of stable bioactive molecules from protein structures with deep learning.

GNINA 1.0: molecular docking with deep learning.

Molecular docking computationally predicts the conformation of a small molecule when binding to a receptor. Scoring functions are a vital piece of any molecular docking pipeline as they determine the fitness of sampled poses. Here we describe and evaluate the 1.0 release of the Gnina docking software, which utilizes an ensemble of convolutional neural networks (CNNs) as a scoring function. We also explore an array of parameter values for Gnina 1.0 to optimize docking performance and computational cost. Docking performance, as evaluated by the percentage of targets where the top pose is better than 2Å root mean square deviation (Top1), is compared to AutoDock Vina scoring when utilizing explicitly defined binding pockets or whole protein docking. GNINA, utilizing a CNN scoring function to rescore the output poses, outperforms AutoDock Vina scoring on redocking and cross-docking tasks when the binding pocket is defined (Top1 increases from 58% to 73% and from 27% to 37%, respectively) and when the whole protein defines the binding pocket (Top1 increases from 31% to 38% and from 12% to 16%, respectively). The derived ensemble of CNNs generalizes to unseen proteins and ligands and produces scores that correlate well with the root mean square deviation to the known binding pose. We provide the 1.0 version of GNINA under an open source license for use as a molecular docking tool at https://github.com/gnina/gnina .

Three-Dimensional Convolutional Neural Networks and a Cross-Docked Data Set for Structure-Based Drug Design.

One of the main challenges in drug discovery is predicting protein-ligand binding affinity. Recently, machine learning approaches have made substantial progress on this task. However, current methods of model evaluation are overly optimistic in measuring generalization to new targets, and there does not exist a standard data set of sufficient size to compare performance between models. We present a new data set for structure-based machine learning, the CrossDocked2020 set, with 22.5 million poses of ligands docked into multiple similar binding pockets across the Protein Data Bank, and perform a comprehensive evaluation of grid-based convolutional neural network (CNN) models on this data set. We also demonstrate how the partitioning of the training data and test data can impact the results of models trained with the PDBbind data set, how performance improves by adding more lower-quality training data, and how training with docked poses imparts pose sensitivity to the predicted affinity of a complex. Our best performing model, an ensemble of five densely connected CNNs, achieves a root mean squared error of 1.42 and Pearson R of 0.612 on the affinity prediction task, an AUC of 0.956 at binding pose classification, and a 68.4% accuracy at pose selection on the CrossDocked2020 set. By providing data splits for clustered cross-validation and the raw data for the CrossDocked2020 set, we establish the first standardized data set for training machine learning models to recognize ligands in noncognate target structures while also greatly expanding the number of poses available for training. In order to facilitate community adoption of this data set for benchmarking protein-ligand binding affinity prediction, we provide our models, weights, and the CrossDocked2020 set at https://github.com/gnina/models.

High-Precision Atomic Charge Prediction for Protein Systems Using Fragment Molecular Orbital Calculation and Machine Learning.

Here, we have constructed neural network-based models that predict atomic partial charges with high accuracy at low computational cost. The models were trained using high-quality data acquired from quantum mechanics calculations using the fragment molecular orbital method. We have succeeded in obtaining highly accurate atomic partial charges for three representative molecular systems of proteins, including one large biomolecule (approx. 2000 atoms). The novelty of our approach is the ability to take into account the electronic polarization in the system, which is a system-dependent phenomenon, being important in the field of drug design. Our high-precision models are useful for the prediction of atomic partial charges and expected to be widely applicable in structure-based drug designs such as structural optimization, high-speed and high-precision docking, and molecular dynamics calculations.

Theoretical Study on the Photoelectron Spectra of Ln(COT)2-: Lanthanide Dependence of the Metal-Ligand Interaction.

We have performed a theoretical analysis of the recently reported photoelectron (PE) spectra of the series of sandwich complex anions Ln(COT)2- (Ln = La-Lu, COT = 1,3,5,7-cyclooctatetraene), focusing on the Ln dependence of the vertical detachment energies. For most Ln, the π molecular orbitals, largely localized on the COT ligands, have the energy order of e1g < e1u < e2g < e2u as in the actinide analogues, reflecting the substantial orbital interaction with the Ln 5d and 5p orbitals. Thus, it would be expected that the lanthanide contraction would increase the orbital interaction so that the overlaps between the COT π and Ln atomic orbitals tend to increase across the series. However, the PE spectra and theoretical calculations were not consistent with this expectation, and the details have been clarified in this study. Furthermore, the energy level splitting patterns of the anion and neutral complexes have been studied by multireference ab initio methods, and the X peak splittings observed in the PE spectra only for the middle-range Ln complexes were found to be due to the specific interaction between the Ln 4f and ligand π orbitals of the neutral complexes in e2u symmetry. Because the magnitude of this 4f-ligand interaction depends critically on the final state 4f electron configuration and the spin state, a significant Ln dependence in the PE spectra is explained.

Dietary administration of probiotics to sows and/or their neonates improves the reproductive performance, incidence of post-weaning diarrhea and histopathological parameters in the intestine of weaned piglets.

Probiotics have gained considerable attention with respect to their beneficial effects on livestock performance and health. The most significant effects of probiotics on the gut microbiota and the host animals take place when they are included in diets during particularly stressful periods such as weaning and/or at the beginning of the lactation period. The probiotics Bacillus mesentericus strain TO-A at 1 × 108 colony forming units (CFU)/g, Clostridium butyricum strain TO-A at 1 × 108 CFU/g and Enterococcus faecalis strain T-110 at 1 × 109 CFU/g were used. Litter weight at delivery and ratio of return to estrous improved significantly (17% and 24% improvement, respectively) by probiotic administration to sows (0.2% (w/w)). Furthermore, the feed intake of the probiotics-administered sows was greater than that of the control sows during the late lactation period. Post-weaning diarrheal incidence and growth performance was improved by probiotics administration to neonates (0.02% (w/w)), while the combined use of probiotics in sows and their neonates induced the enlargement of villous height and prevented muscle layer thinning in the small intestine of weaning piglets. The administration of probiotics of three species of live bacteria improved the porcine reproductive performance around stressful periods of sows (farrowing) and piglets (weaning). [Corrections added on 26 April 2016, after first online publication: 'Enterococcus faecalis strain T-100' has been corrected to 'Enterococcus faecalis strain T-110' in the above paragraph and in the 'Probiotics' section under the Materials and Methods heading.].

Investigation of the electronic structures of organolanthanide sandwich complex anions by photoelectron spectroscopy: 4f orbital contribution in the metal-ligand interaction.

The electronic structures of lanthanide (Ln) ions sandwiched between 1,3,5,7-cyclooctatetraene (COT), Ln(COT)2(-), have been investigated by anion photoelectron spectroscopy. Complexes of 12 Ln atoms were investigated (excluding promethium (Pm), europium (Eu), and ytterbium (Yb)). The 213 nm photoelectron (PE) spectra of Ln(COT)2(-) exhibit two peaks assignable to the highest occupied molecular orbital (HOMO; e2u) and the next HOMO (HOMO-1; e2g) approximately at 2.6 and 3.6 eV, respectively, and their energy gap increases as the central metal atom progresses from lanthanum (La) to lutetium (Lu). Since lanthanide contraction shortens the distance between the Ln atom and the COT ligands, the widening energy gap represents the destabilization of the e2u orbital as well as the stabilization of the e2g orbital. Evidence for 4f orbital contribution in the metal-ligand interaction has been revealed by the Ln atom dependence in which the same e2u orbital symmetry enables an interaction between the 4f orbital of Ln atoms and the π orbital of COT.