Machine learning (ML) has become a central focus of the computational chemistry community. I will first discuss my personal history in the field. Then I will provide a broader view of how this resurgence in ML interest echoes and advances upon earlier efforts. Although numerous changes have brought about this latest wave, one of the most significant is the increased accuracy and efficiency of low‐cost methods (e. g., density functional theory or DFT) that have made it possible to generate large data sets for ML models. ML has also been used to bypass, guide, or improve DFT. The field of computational chemistry thus finds itself at a crossroads as ML both augments and supersedes traditional efforts. I will present what I believe the role of the computational chemist will be in this evolving landscape, with specific focus on my experience in the development of autonomous workflows in computational materials discovery for open‐shell transition‐metal chemistry.
As machine learning (ML) has matured, it has opened a new frontier in theoretical and computational chemistry by offering the promise of simultaneous paradigm shifts in accuracy and efficiency. Nowhere is this advance more needed, but also more challenging to achieve, than in the discovery of open‐shell transition metal complexes. Here, localized
This article is categorized under:
Electronic Structure Theory > Density Functional Theory
Software > Molecular Modeling
Computer and Information Science > Chemoinformatics
more » « less- NSF-PAR ID:
- 10360778
- Publisher / Repository:
- Wiley Blackwell (John Wiley & Sons)
- Date Published:
- Journal Name:
- WIREs Computational Molecular Science
- Volume:
- 10
- Issue:
- 1
- ISSN:
- 1759-0876
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract -
Chiral transition-metal complexes are of interest in many fields ranging from asymmetric catalysis and molecular materials science to optoelectronic applications or fundamental physics including parity violation effects. We present here a combined theoretical and experimental investigation of gas-phase valence-shell photoelectron circular dichroism (PECD) on the challenging open-shell ruthenium( iii )-tris-(acetylacetonato) complex, Ru(acac) 3 . Enantiomerically pure Δ- or Λ-Ru(acac) 3 , characterized by electronic circular dichroism (ECD), were vaporized and adiabatically expanded to produce a supersonic beam and photoionized by circularly-polarized VUV light from the DESIRS beamline at Synchrotron SOLEIL. Photoelectron spectroscopy (PES) and PECD experiments were conducted using a double imaging electron/ion coincidence spectrometer, and compared to density functional theory (DFT) and time-dependent DFT (TDDFT) calculations. The open-shell character of Ru(acac) 3 , which is not taken into account in our DFT approach, is expected to give rise to a wide multiplet structure, which is not resolved in our PES signals but whose presence might be inferred from the additional striking features observed in the PECD curves. Nevertheless, the DFT-based assignment of the electronic bands leads to the characterisation of the ionized orbitals. In line with other recent works, the results confirm that PECD persists independently on the localization and/or on the achiral or chiral nature of the initial orbital, but is rather a probe of the molecular potential as a whole. Overall, the measured PECD signals on Ru(acac) 3 , a system exhibiting D 3 propeller-type chirality, are of similar magnitude compared to those on asymmetric-carbon-based chiral organic molecules which constitute the vast majority of species investigated so far, thus suggesting that PECD is a universal mechanism, inherent to any type of chirality.more » « less
-
Machine learning (ML) offers an attractive method for making predictions about molecular systems while circumventing the need to run expensive electronic structure calculations. Once trained on ab initio data, the promise of ML is to deliver accurate predictions of molecular properties that were previously computationally infeasible. In this work, we develop and train a graph neural network model to correct the basis set incompleteness error (BSIE) between a small and large basis set at the RHF and B3LYP levels of theory. Our results show that, when compared to fitting to the total potential, an ML model fitted to correct the BSIE is better at generalizing to systems not seen during training. We test this ability by training on single molecules while evaluating on molecular complexes. We also show that ensemble models yield better behaved potentials in situations where the training data is insufficient. However, even when only fitting to the BSIE, acceptable performance is only achieved when the training data sufficiently resemble the systems one wants to make predictions on. The test error of the final model trained to predict the difference between the cc-pVDZ and cc-pV5Z potential is 0.184 kcal/mol for the B3LYP density functional, and the ensemble model accurately reproduces the large basis set interaction energy curves on the S66x8 dataset.more » « less
-
more » « less
Abstract Argumentation, a key scientific practice presented in the
Framework for K-12 Science Education , requires students to construct and critique arguments, but timely evaluation of arguments in large-scale classrooms is challenging. Recent work has shown the potential of automated scoring systems for open response assessments, leveraging machine learning (ML) and artificial intelligence (AI) to aid the scoring of written arguments in complex assessments. Moreover, research has amplified that the features (i.e., complexity, diversity, and structure) of assessment construct are critical to ML scoring accuracy, yet how the assessment construct may be associated with machine scoring accuracy remains unknown. This study investigated how the features associated with the assessment construct of a scientific argumentation assessment item affected machine scoring performance. Specifically, we conceptualized the construct in three dimensions: complexity, diversity, and structure. We employed human experts to code characteristics of the assessment tasks and score middle school student responses to 17 argumentation tasks aligned to three levels of a validated learning progression of scientific argumentation. We randomly selected 361 responses to use as training sets to build machine-learning scoring models for each item. The scoring models yielded a range of agreements with human consensus scores, measured by Cohen’s kappa (mean = 0.60; range 0.38 − 0.89), indicating good to almost perfect performance. We found that higher levels ofComplexity andDiversity of the assessment task were associated with decreased model performance, similarly the relationship between levels ofStructure and model performance showed a somewhat negative linear trend. These findings highlight the importance of considering these construct characteristics when developing ML models for scoring assessments, particularly for higher complexity items and multidimensional assessments. -
Modern semiempirical electronic structure methods have considerable promise in drug discovery as universal “force fields” that can reliably model biological and drug-like molecules, including alternative tautomers and protonation states. Herein, we compare the performance of several neglect of diatomic differential overlap-based semiempirical (MNDO/d, AM1, PM6, PM6-D3H4X, PM7, and ODM2), density-functional tight-binding based (DFTB3, DFTB/ChIMES, GFN1-xTB, and GFN2-xTB) models with pure machine learning potentials (ANI-1x and ANI-2x) and hybrid quantum mechanical/machine learning potentials (AIQM1 and QD π) for a wide range of data computed at a consistent ωB97X/6-31G* level of theory (as in the ANI-1x database). This data includes conformational energies, intermolecular interactions, tautomers, and protonation states. Additional comparisons are made to a set of natural and synthetic nucleic acids from the artificially expanded genetic information system that has important implications for the design of new biotechnology and therapeutics. Finally, we examine the acid/base chemistry relevant for RNA cleavage reactions catalyzed by small nucleolytic ribozymes, DNAzymes, and ribonucleases. Overall, the hybrid quantum mechanical/machine learning potentials appear to be the most robust for these datasets, and the recently developed QD π model performs exceptionally well, having especially high accuracy for tautomers and protonation states relevant to drug discovery.more » « less
