Predictive Minisci late stage functionalization with transfer learning.

Structural diversification of lead molecules is a key component of drug discovery to explore chemical space. Late-stage functionalizations (LSFs) are versatile methodologies capable of installing functional handles on richly decorated intermediates to deliver numerous diverse products in a single reaction. Predicting the regioselectivity of LSF is still an open challenge in the field. Numerous efforts from chemoinformatics and machine learning (ML) groups have made strides in this area. However, it is arduous to isolate and characterize the multitude of LSF products generated, limiting available data and hindering pure ML approaches. We report the development of an approach that combines a message passing neural network and 13C NMR-based transfer learning to predict the atom-wise probabilities of functionalization for Minisci and P450-based functionalizations. We validated our model both retrospectively and with a series of prospective experiments, showing that it accurately predicts the outcomes of Minisci-type and P450 transformations and outperforms the well-established Fukui-based reactivity indices and other machine learning reactivity-based algorithms.

A Non-Isothermal Pore Network Model of Primary Freeze Drying.

In this work, a non-isothermal pore network (PN) model with quasi-steady vapor transport and transient heat transfer is presented for the first time for the application of primary freeze drying. The pore-scale resolved model is physically based and allows for the investigation of correlations between spatially distributed structure and transport conditions. The studied examples were regular PN lattices with a significantly different structure, namely a spatially homogeneous PN, also denoted as monomodal PN, and a PN with significant structure variation, referred to as bimodal PN because of its bimodal pore size distribution. The material properties selected for the solid skeleton in this study are equivalent to those of maltodextrin. The temperature ranges applied here were -28 °C to -18 °C in the PN and -42 °C in the surrounding environment. The environmental vapor pressure was 10 Pa. The PNs were dried with constant temperature boundary conditions, and heat was transferred at the top side by the vapor leaving the PN. It is shown how the structural peculiarities affect the local heat and mass transfer conditions and result in a significant widening of the sublimation front in the case of the bimodal PN. The possibility of spatially and temporally resolved front structures is a unique feature of the PN model and allows the study of situations that are not yet described by classical continuum approaches, namely heterogeneous frozen porous materials. As demonstrated by the thin layers studied here, the pore-scale simulations are of particular interest for such situations, such as in lyomicroscopes or collagen scaffolds, where a length-scale separation between dry and ice-saturated regions is not possible.

GPU-accelerated approximate kernel method for quantum machine learning.

We introduce Quantum Machine Learning (QML)-Lightning, a PyTorch package containing graphics processing unit (GPU)-accelerated approximate kernel models, which can yield trained models within seconds. QML-Lightning includes a cost-efficient GPU implementation of FCHL19, which together can provide energy and force predictions with competitive accuracy on a microsecond per atom timescale. Using modern GPU hardware, we report learning curves of energies and forces as well as timings as numerical evidence for select legacy benchmarks from atomistic simulation including QM9, MD-17, and 3BPA.

Rapid discovery of stable materials by coordinate-free coarse graining.

A fundamental challenge in materials science pertains to elucidating the relationship between stoichiometry, stability, structure, and property. Recent advances have shown that machine learning can be used to learn such relationships, allowing the stability and functional properties of materials to be accurately predicted. However, most of these approaches use atomic coordinates as input and are thus bottlenecked by crystal structure identification when investigating previously unidentified materials. Our approach solves this bottleneck by coarse-graining the infinite search space of atomic coordinates into a combinatorially enumerable search space. The key idea is to use Wyckoff representations, coordinate-free sets of symmetry-related positions in a crystal, as the input to a machine learning model. Our model demonstrates exceptionally high precision in finding unknown theoretically stable materials, identifying 1569 materials that lie below the known convex hull of previously calculated materials from just 5675 ab initio calculations. Our approach opens up fundamental advances in computational materials discovery.

Cardiorespiratory kinetics in exercise physiology: estimates and predictions using randomized changes in work rate.

PURPOSE: Kinetics of cardiorespiratory parameters (CRP) in response to work rate (WR) changes are evaluated by pseudo-random binary sequences (PRBS testing). In this study, two algorithms were applied to convert responses from PRBS testing into appropriate impulse responses to predict steady states values and responses to incremental increases in exercise intensity. METHODS: 13 individuals (age: 41 ± 9 years, BMI: 23.8 ± 3.7 kg m-2), completing an exercise test protocol, comprising a section of randomized changes of 30 W and 80 W (PRBS), two phases of constant WR at 30 W and 80 W and incremental WR until subjective fatigue, were included in the analysis. Ventilation ([Formula: see text]), O2 uptake ([Formula: see text]), CO2 output ([Formula: see text]) and heart rate (HR) were monitored. Impulse responses were calculated in the time domain and in the frequency domain from the cross-correlations of WR and the respective CRP. RESULTS: The algorithm in the time domain allows better prediction for [Formula: see text] and [Formula: see text], whereas for [Formula: see text] and HR the results were similar for both algorithms. Best predictions were found for [Formula: see text] and HR with higher (3-4%) 30 W steady states and lower (1-4%) values for 80 W. Tendencies were found in the residuals between predicted and measured data. CONCLUSION: The CRP kinetics, resulting from PRBS testing, are qualified to assess steady states within the applied WR range. Below the ventilatory threshold, [Formula: see text] and HR responses to incrementally increasing exercise intensities can be sufficiently predicted.

FCHL revisited: Faster and more accurate quantum machine learning.

We introduce the FCHL19 representation for atomic environments in molecules or condensed-phase systems. Machine learning models based on FCHL19 are able to yield predictions of atomic forces and energies of query compounds with chemical accuracy on the scale of milliseconds. FCHL19 is a revision of our previous work [F. A. Faber et al., J. Chem. Phys. 148, 241717 (2018)] where the representation is discretized and the individual features are rigorously optimized using Monte Carlo optimization. Combined with a Gaussian kernel function that incorporates elemental screening, chemical accuracy is reached for energy learning on the QM7b and QM9 datasets after training for minutes and hours, respectively. The model also shows good performance for non-bonded interactions in the condensed phase for a set of water clusters with a mean absolute error (MAE) binding energy error of less than 0.1 kcal/mol/molecule after training on 3200 samples. For force learning on the MD17 dataset, our optimized model similarly displays state-of-the-art accuracy with a regressor based on Gaussian process regression. When the revised FCHL19 representation is combined with the operator quantum machine learning regressor, forces and energies can be predicted in only a few milliseconds per atom. The model presented herein is fast and lightweight enough for use in general chemistry problems as well as molecular dynamics simulations.

Operators in quantum machine learning: Response properties in chemical space.

The role of response operators is well established in quantum mechanics. We investigate their use for universal quantum machine learning models of response properties in molecules. After introducing a theoretical basis, we present and discuss numerical evidence based on measuring the potential energy's response with respect to atomic displacement and to electric fields. Prediction errors for corresponding properties, atomic forces, and dipole moments improve in a systematic fashion with training set size and reach high accuracy for small training sets. Prediction of normal modes and infrared-spectra of some small molecules demonstrates the usefulness of this approach for chemistry.

Alchemical and structural distribution based representation for universal quantum machine learning.

We introduce a representation of any atom in any chemical environment for the automatized generation of universal kernel ridge regression-based quantum machine learning (QML) models of electronic properties, trained throughout chemical compound space. The representation is based on Gaussian distribution functions, scaled by power laws and explicitly accounting for structural as well as elemental degrees of freedom. The elemental components help us to lower the QML model's learning curve, and, through interpolation across the periodic table, even enable "alchemical extrapolation" to covalent bonding between elements not part of training. This point is demonstrated for the prediction of covalent binding in single, double, and triple bonds among main-group elements as well as for atomization energies in organic molecules. We present numerical evidence that resulting QML energy models, after training on a few thousand random training instances, reach chemical accuracy for out-of-sample compounds. Compound datasets studied include thousands of structurally and compositionally diverse organic molecules, non-covalently bonded protein side-chains, (H2O)40-clusters, and crystalline solids. Learning curves for QML models also indicate competitive predictive power for various other electronic ground state properties of organic molecules, calculated with hybrid density functional theory, including polarizability, heat-capacity, HOMO-LUMO eigenvalues and gap, zero point vibrational energy, dipole moment, and highest vibrational fundamental frequency.

Prediction Errors of Molecular Machine Learning Models Lower than Hybrid DFT Error.

We investigate the impact of choosing regressors and molecular representations for the construction of fast machine learning (ML) models of 13 electronic ground-state properties of organic molecules. The performance of each regressor/representation/property combination is assessed using learning curves which report out-of-sample errors as a function of training set size with up to ∼118k distinct molecules. Molecular structures and properties at the hybrid density functional theory (DFT) level of theory come from the QM9 database [ Ramakrishnan et al. Sci. Data 2014 , 1 , 140022 ] and include enthalpies and free energies of atomization, HOMO/LUMO energies and gap, dipole moment, polarizability, zero point vibrational energy, heat capacity, and the highest fundamental vibrational frequency. Various molecular representations have been studied (Coulomb matrix, bag of bonds, BAML and ECFP4, molecular graphs (MG)), as well as newly developed distribution based variants including histograms of distances (HD), angles (HDA/MARAD), and dihedrals (HDAD). Regressors include linear models (Bayesian ridge regression (BR) and linear regression with elastic net regularization (EN)), random forest (RF), kernel ridge regression (KRR), and two types of neural networks, graph convolutions (GC) and gated graph networks (GG). Out-of sample errors are strongly dependent on the choice of representation and regressor and molecular property. Electronic properties are typically best accounted for by MG and GC, while energetic properties are better described by HDAD and KRR. The specific combinations with the lowest out-of-sample errors in the ∼118k training set size limit are (free) energies and enthalpies of atomization (HDAD/KRR), HOMO/LUMO eigenvalue and gap (MG/GC), dipole moment (MG/GC), static polarizability (MG/GG), zero point vibrational energy (HDAD/KRR), heat capacity at room temperature (HDAD/KRR), and highest fundamental vibrational frequency (BAML/RF). We present numerical evidence that ML model predictions deviate from DFT (B3LYP) less than DFT (B3LYP) deviates from experiment for all properties. Furthermore, out-of-sample prediction errors with respect to hybrid DFT reference are on par with, or close to, chemical accuracy. The results suggest that ML models could be more accurate than hybrid DFT if explicitly electron correlated quantum (or experimental) data were available.

Machine Learning Energies of 2 Million Elpasolite (ABC_{2}D_{6}) Crystals.

Elpasolite is the predominant quaternary crystal structure (AlNaK_{2}F_{6} prototype) reported in the Inorganic Crystal Structure Database. We develop a machine learning model to calculate density functional theory quality formation energies of all ∼2×10^{6} pristine ABC_{2}D_{6} elpasolite crystals that can be made up from main-group elements (up to bismuth). Our model's accuracy can be improved systematically, reaching a mean absolute error of 0.1 eV/atom for a training set consisting of 10×10^{3} crystals. Important bonding trends are revealed: fluoride is best suited to fit the coordination of the D site, which lowers the formation energy whereas the opposite is found for carbon. The bonding contribution of the elements A and B is very small on average. Low formation energies result from A and B being late elements from group II, C being a late (group I) element, and D being fluoride. Out of 2×10^{6} crystals, 90 unique structures are predicted to be on the convex hull-among which is NFAl_{2}Ca_{6}, with a peculiar stoichiometry and a negative atomic oxidation state for Al.