OrbNet Denali: A machine learning potential for biological and organic chemistry with semi-empirical cost and DFT accuracy.

By Anders S. Christensen, Sai Krishna Sirumalla, Zhuoran Qiao, M. O’Connor, Daniel G. A. Smith, Feizhi Ding, P. J. Bygrave, Anima Anandkumar, Matthew Welborn, Frederick R. Manby and 1 more. 2021

87 citations recorded by the source corpus.

We present OrbNet Denali, a machine learning model for an electronic structure that is designed as a drop-in replacement for ground-state density functional theory (DFT) energy calculations. The model is a message-passing graph neural network that uses symmetry-adapted atomic orbital features from a low-cost quantum calculation to predict the energy of a molecule. OrbNet Denali is trained on a vast dataset of 2.3 × 106 DFT calculations on molecules and geometries. This dataset covers the most common elements in biochemistry and organic chemistry (H, Li, B, C, N, O, F, Na, Mg, Si, P, S, Cl, K, Ca, Br, and I) and charged molecules. OrbNet Denali is demonstrated on several well-established benchmark datasets, and we find that it provides accuracy that is on par with modern DFT methods while offering a speedup of up to three orders of magnitude. For the GMTKN55 benchmark set, OrbNet Denali achieves WTMAD-1 and WTMAD-2 scores of 7.19 and 9.84, on par with modern DFT functionals. For several GMTKN55 subsets, which contain chemical problems that are not present in the training set, OrbNet Denali produces a mean absolute error comparable to those of DFT methods. For the Hutchison conformer benchmark set, OrbNet Denali has a median correlation coefficient of R2 = 0.90 compared to the reference DLPNO-CCSD(T) calculation and R2 = 0.97 compared to the method used to generate the training data (ωB97X-D3/def2-TZVP), exceeding the performance of any other method with a similar cost. Similarly, the model reaches chemical accuracy for non-covalent interactions in the S66x10 dataset. For torsional profiles, OrbNet Denali reproduces the torsion profiles of ωB97X-D3/def2-TZVP with an average mean absolute error of 0.12 kcal/mol for the potential energy surfaces of the diverse fragments in the TorsionNet500 dataset.

Subfields: Inorganic Chemistry, Materials Chemistry, Physical and Theoretical Chemistry

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Materials Chemistry

1. Kenneth Atz
2. Gisbert Schneider
3. José Jiménez-Luna
4. Clemens Isert
5. Francesca Grisoni

In Physical and Theoretical Chemistry

1. Gisbert Schneider
2. José Jiménez-Luna
3. Clemens Isert
4. Kenneth Atz

Citations in the corpus

• QMugs, quantum mechanical properties of drug-like molecules (2022, methodology): “Similar to previous studies 13,43 , the semiempirical GFN2-xTB method was used as a surrogate for full DFT geometry optimization, as the latter is associated with challenging computational costs when considering the size and number of molecules in the QMugs dataset.”