Quantum chemistry structures and properties of 134 kilo molecules
By R. Ramakrishnan, Pavlo O. Dral, Matthias Rupp , O. V. Lilienfeld. 2014
2,384 citations recorded by the source corpus.
Computational de novo design of new drugs and materials requires rigorous and unbiased exploration of chemical compound space. However, large uncharted territories persist due to its size scaling combinatorially with molecular size. We report computed geometric, energetic, electronic, and thermodynamic properties for 134k stable small organic molecules made up of CHONF. These molecules correspond to the subset of all 133,885 species with up to nine heavy atoms (CONF) out of the GDB-17 chemical universe of 166 billion organic molecules. We report geometries minimal in energy, corresponding harmonic frequencies, dipole moments, polarizabilities, along with energies, enthalpies, and free energies of atomization. All properties were calculated at the B3LYP/6-31G(2df,p) level of quantum chemistry. Furthermore, for the predominant stoichiometry, C7H10O2, there are 6,095 constitutional isomers among the 134k molecules. We report energies, enthalpies, and free energies of atomization at the more accurate G4MP2 level of theory for all of them. As such, this data set provides quantum chemical properties for a relevant, consistent, and comprehensive chemical space of small organic molecules. This database may serve the benchmarking of existing methods, development of new methods, such as hybrid quantum mechanics/machine learning, and systematic identification of structure-property relationships. Design Type(s) in silico design • data integration Measurement Type(s) Computational Chemistry Technology Type(s) quantum chemistry computational method Factor Type(s) level of theory Design Type(s) in silico design • data integration Measurement Type(s) Computational Chemistry Technology Type(s) quantum chemistry computational method Factor Type(s) level of theory Machine-accessible metadata file describing the reported data (ISA-Tab format)
Subfields: Atomic and Molecular Physics, and Optics, Computational Theory and Mathematics, Organic Chemistry, Computational Chemistry & Drug Discovery
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Atomic and Molecular Physics, and Optics
- K. Müller
- Alán Aspuru-Guzik
- A. Pais
- Tânia F Cova
- Alexandre Tkatchenko
- Stéphane Mallat
- Matthew J. Hirn
- Tess E. Smidt
In Computational Theory and Mathematics
- Tommi S. Jaakkola
- Regina Barzilay
- Tess E. Smidt
- Max Welling
- Jian Tang
- William L. Hamilton
- Jure Leskovec
- Gisbert Schneider
In Organic Chemistry
- Alán Aspuru-Guzik
- Jian Tang
- Eran Yahav
- Uri Alon
- Tess E. Smidt
- Katsuhiko Ishiguro
- Masanori Koyama
- Shin-ichi Maeda
In Computational Chemistry & Drug Discovery
- Tess E. Smidt
- K. Müller
- Tommi S. Jaakkola
- Regina Barzilay
- Grgoire Montavon
- Risi Kondor
- Kristof T. Schütt
- Alexandre Tkatchenko
Citations in the corpus
- Neural Message Passing for Quantum Chemistry (2017, methodology): “Here we focus on the QM9 dataset as such a benchmark (Ramakrishnan et al., 2014).”
- E(n) Equivariant Graph Neural Networks (2021, methodology): “For QM9 (Ramakrishnan et al., 2014) we used the dataset partitions from (Anderson et al., 2019).”
- Directional Message Passing for Molecular Graphs (2020, methodology): “We test DimeNet’s performance for predicting molecular properties using the common QM9 benchmark (Ramakrishnan et al., 2014).”
- Fast Graph Representation Learning with PyTorch Geometric (2019, background): “In detail, we provide over 60 graph kernel benchmark datasets 3 (Kersting et al., 2016), e.g. , PROTEINS or IMDB-BINARY, the citation graphs Cora, CiteSeer, PubMed and Cora-Full (Sen et al., 2008; Bojchevski & Günnemann, 2018), the Coauthor CS/Physics and Amazon Computers/Photo datasets from Shchur ”
- Graph Warp Module: an Auxiliary Module for Boosting the Power of Graph Neural Networks (2019, methodology): “1 Datasets For the regression tasks, we used the QM9 dataset [20, 19] and the Lipophilicity (LIPO) dataset for the graph regression task.”
- SchNet: A continuous-filter convolutional neural network for modeling quantum interactions (2017, background): “Data sets such as the established QM9 benchmark [8] contain only equilibrium molecules.”
- Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules (2016, neutral): “Two autoencoder systems were trained: one with 108 000 molecules from the QM9 data set of molecules with fewer than 9 heavy atoms 31 and another with 250 000 drug-like commercially available molecules extracted at random from the ZINC database.”
- Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties (2020, methodology): “We explore this question using the QM9 benchmark [9, 10] by predicting quantum chemical properties of small molecules.”
- SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks (2020, background): “Point cloud data is ubiquitous across many fields, presenting itself in diverse forms such as 3D object scans [21], molecular structures [17], or particle simulations [10].”
- Cormorant: Covariant Molecular Neural Networks (2019, background): “QM9 [Ramakrishnan et al., 2014] is a dataset of approximately 134 k small organic molecules containing the atoms H, C, N, O, F.”
- Tensor Field Networks: Rotation- and Translation-Equivariant Neural Networks for 3D Point Clouds (2018, methodology): “We trained on structures in the QM9 dataset, a collection of 134,000 molecules with up to nine heavy atoms (C, N, O, F); including hydrogens, there are up to 29 atoms per molecule (Ramakrishnan et al., 2014).”
- GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders (2018, positive): “QM9 dataset (Ramakrishnan et al., 2014) contains about 134k organic molecules of up to 9 heavy (non Hydrogen) atoms with 4 distinct atomic numbers and 4 bond types, we set k = 9 , d e = 4 and d n = 4 .”
- Provably Powerful Graph Networks (2019, methodology): “For the regression task, we conducted an experiment on a standard graph learning benchmark called the QM9 dataset (Ramakrishnan et al., 2014; Wu et al., 2018).”