Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userIMPRESSION -- Prediction of NMR Parameters for 3-dimensional chemical structures using Machine Learning with near quantum chemical accuracy
62
0
0.0
(
0
)
Added by
Lars Andersen Bratholm
Publication date
2019
fields
Physics
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
The IMPRESSION (Intelligent Machine PREdiction of Shift and Scalar Information Of Nuclei) machine learning system provides an efficient and accurate route to the prediction of NMR parameters from 3-dimensional chemical structures. Here we demonstrate that machine learning predictions, trained on quantum chemical computed values for NMR parameters, are essentially as accurate but computationally much more efficient (tens of milliseconds per molecule) than quantum chemical calculations (hours/days per molecule). Training the machine learning systems on quantum chemical, rather than experimental, data circumvents the need for existence of large, structurally diverse, error-free experimental databases and makes IMPRESSION applicable to solving 3-dimensional problems such as molecular conformation and isomerism
rate research
Read More
Machine learning techniques applied to chemical reactions has a long history. The present contribution discusses applications ranging from small molecule reaction dynamics to platforms for reaction planning. ML-based techniques can be of particular interest for problems which involve both, computation and experiments. For one, Bayesian inference is a powerful approach to include knowledge from experiment in improving computational models. ML-based methods can also be used to handle problems that are formally intractable using conventional approaches, such as exhaustive characterization of state-to-state information in reactive collisions. Finally, the explicit simulation of reactive networks as they occur in combustion has become possible using machine-learned neural network potentials. This review provides an overview of the questions that can and have been addressed using machine learning techniques and an outlook discusses challenges in this diverse and stimulating field. It is concluded that ML applied to chemistry problems as practiced and conceived today has the potential to transform the way with which the field approaches problems involving chemical reactions, both, in research and academic teaching.
We report an accurate study of interactions between Benzene molecules using variational quantum Monte Carlo (VMC) and diffusion quantum Monte Carlo (DMC) methods. We compare these results with density functional theory (DFT) using different van der Waals (vdW) functionals. In our QMC calculations, we use accurate correlated trial wave functions including three-body Jastrow factors, and backflow transformations. We consider two benzene molecules in the parallel displaced (PD) geometry, and find that by highly optimizing the wave function and introducing more dynamical correlation into the wave function, we compute the weak chemical binding energy between aromatic rings accurately. We find optimal VMC and DMC binding energies of -2.3(4) and -2.7(3) kcal/mol, respectively. The best estimate of the CCSD(T)/CBS limit is -2.65(2) kcal/mol [E. Miliordos et al, J. Phys. Chem. A 118, 7568 (2014)]. Our results indicate that QMC methods give chemical accuracy for weakly bound van der Waals molecular interactions, comparable to results from the best quantum chemistry methods.
Strategies for machine-learning(ML)-accelerated discovery that are general across materials composition spaces are essential, but demonstrations of ML have been primarily limited to narrow composition variations. By addressing the scarcity of data in promising regions of chemical space for challenging targets like open-shell transition-metal complexes, general representations and transferable ML models that leverage known relationships in existing data will accelerate discovery. Over a large set (ca. 1000) of isovalent transition-metal complexes, we quantify evident relationships for different properties (i.e., spin-splitting and ligand dissociation) between rows of the periodic table (i.e., 3d/4d metals and 2p/3p ligands). We demonstrate an extension to graph-based revised autocorrelation (RAC) representation (i.e., eRAC) that incorporates the effective nuclear charge alongside the nuclear charge heuristic that otherwise overestimates dissimilarity of isovalent complexes. To address the common challenge of discovery in a new space where data is limited, we introduce a transfer learning approach in which we seed models trained on a large amount of data from one row of the periodic table with a small number of data points from the additional row. We demonstrate the synergistic value of the eRACs alongside this transfer learning strategy to consistently improve model performance. Analysis of these models highlights how the approach succeeds by reordering the distances between complexes to be more consistent with the periodic table, a property we expect to be broadly useful for other materials domains.
We present a scheme to obtain an inexpensive and reliable estimate of the uncertainty associated with the predictions of a machine-learning model of atomic and molecular properties. The scheme is based on resampling, with multiple models being generated based on sub-sampling of the same training data. The accuracy of the uncertainty prediction can be benchmarked by maximum likelihood estimation, which can also be used to correct for correlations between resampled models, and to improve the performance of the uncertainty estimation by a cross-validation procedure. In the case of sparse Gaussian Process Regression models, this resampled estimator can be evaluated at negligible cost. We demonstrate the reliability of these estimates for the prediction of molecular energetics, and for the estimation of nuclear chemical shieldings in molecular crystals. Extension to estimate the uncertainty in energy differences, forces, or other correlated predictions is straightforward. This method can be easily applied to other machine learning schemes, and will be beneficial to make data-driven predictions more reliable, and to facilitate training-set optimization and active-learning strategies.
The requirement for accelerated and quantitatively accurate screening of nuclear magnetic resonance spectra across the small molecules chemical compound space is two-fold: (1) a robust `local machine learning (ML) strategy capturing the effect of neighbourhood on an atoms `near-sighted property -- chemical shielding; (2) an accurate reference dataset generated with a state-of-the-art first principles method for training. Herein we report the QM9-NMR dataset comprising isotropic shielding of over 0.8 million C atoms in 134k molecules of the QM9 dataset in gas and five common solvent phases. Using these data for training, we present benchmark results for the prediction transferability of kernel-ridge regression models with popular local descriptors. Our best model trained on 100k samples, accurately predict isotropic shielding of 50k `hold-out atoms with a mean error of less than $1.9$ ppm. For rapid prediction of new query molecules, the models were trained on geometries from an inexpensive theory. Furthermore, by using a $Delta$-ML strategy, we quench the error below $1.4$ ppm. Finally, we test the transferability on non-trivial benchmark sets that include benchmark molecules comprising 10 to 17 heavy atoms and drugs.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
