CoChat
Sign Up
Cited by 161Machine Learning in Materials ScienceAdvanced Chemical Physics StudiesComputational Drug Discovery Methods

Strategies and Software for Machine Learning Accelerated Discovery in Transition Metal Chemistry

Open access
Nandy, Aditya, Duan, Chenru, Janet, Jon Paul, Gugler, Stefan, Kulik, Heather J.·2018·Industrial & Engineering Chemistry Research
DOI 10.1021/acs.iecr.8b04015

Published version of a preprint

Open this paper in CoChat — free account

Abstract

Machine learning the electronic structure of open shell transition metal complexes presents unique challenges, including robust and automated data set generation. Here, we introduce tools that simplify data acquisition from density functional theory (DFT) and validation of trained machine learning models using the molSimplify automatic design (mAD) workflow. We demonstrate this workflow by training and comparing the performance of LASSO, kernel ridge regression (KRR), and artificial neural network (ANN) models using heuristic, topological revised autocorrelation (RAC) descriptors we have recently introduced for machine learning inorganic chemistry. On a series of open shell transition metal complexes, we evaluate set aside test errors of these models for predicting the HOMO level and HOMO-LUMO gap. The best performing models are ANNs, which show 0.15 and 0.25 eV test set mean absolute errors on the HOMO level and HOMO-LUMO gap, respectively. Poor performing KRR models using the full 153-feature RAC set are improved to nearly the same performance as the ANNs when trained on down-selected subsets of 20-30 features. Analysis of the essential descriptors for HOMO and HOMO-LUMO gap prediction as well as comparison to subsets previously obtained for other properties reveals the paramount importance of non-local, steric properties in determining frontier molecular orbital energetics. We demonstrate our model performance on diverse complexes and in the discovery of molecules with target HOMO-LUMO gaps from a large 15,000 molecule design space in minutes rather than days that full DFT evaluation would require.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1021/acs.iecr.8b04015

Cites

98 of 98 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Identifying key descriptors or features in a structure-property relationship mapping has been demonstrated as an approach to revealing design principles and enabling iterative design.

    “Identifying key descriptors or features in a structure-property relationship mapping has been demonstrated as an approach to revealing design principles (18,22,32,52) and enabling iterative design (70).”

    From Machine Learning for the Discovery, Design, and Engineering of Materials. (2022) · cites this paper as [22]

    Section 4a (page unavailable) says feature selection identifies the relative importance of descriptors, while Section 2b (page unavailable) says analyzed results guide iterative discovery in mAD. The paper does not establish that identifying key descriptors itself enables that iterative design; it presents these as separate parts of the workflow. (LLM, conf=0.82)

    Experimental — accuracy not yet measured

  • In some cases, deep learning models have been developed to predict frontier orbital energies, which can be applied to explain experimental activity of catalysts known to be correlated to such energies.

    “In some cases, deep learning models (e.g., ANNs) have been developed to predict frontier orbital energies (22,43,86), which can be applied to explain experimental activity of catalysts known to be correlated to such energies (86).”

    From Machine Learning for the Discovery, Design, and Engineering of Materials. (2022) · cites this paper as [22]

    The Abstract and Section 5 (page numbers unavailable in our copy) report ANN models that predict HOMO levels and HOMO–LUMO gaps in transition-metal complexes. They do not report using those predictions to explain experimental catalyst activity or establish a correlation between that activity and frontier-orbital energies. (LLM, conf=0.86)

    Experimental — accuracy not yet measured

  • By applying random forest-ranked recursive feature addition to KRR models, expected heuristics have been reinforced by feature selection approaches, such as the spin splitting of a complex being strongly dependent on the metal-local electronic environment.

    “By applying random forest-ranked recursive feature addition to KRR models, expected heuristics have been reinforced by feature selection approaches (22,32,105), such as the spin splitting of a complex being strongly dependent on the metal-local electronic environment (e.g., from the electronegativities of surrounding atoms; Figure 5)”

    From Machine Learning for the Discovery, Design, and Engineering of Materials. (2022) · cites this paper as [22]

    Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

    Experimental — accuracy not yet measured

  • One-shot models such as LASSO or random forests are widely used both as ML models for property prediction and for selection of important features.

    “In a similar fashion, one-shot models such as least absolute shrinkage and selection operator (LASSO) or random forests are widely used both as ML models for property prediction (61) and for selection of important features (22,32,40,43) (Figure 4).”

    From Machine Learning for the Discovery, Design, and Engineering of Materials. (2022) · cites this paper as [22]

  • When KRR models are applied with tailored feature sets, their prediction accuracy can rival or surpass deep learning models.

    “When KRR models are applied with tailored feature sets, their prediction accuracy can rival or surpass (22) deep learning models (e.g., fully connected ANNs or convolutional neural networks).”

    From Machine Learning for the Discovery, Design, and Engineering of Materials. (2022) · cites this paper as [22]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Nandy, A., Duan, C., Janet, J. P., Gugler, S., & Kulik, H. J.
Affiliations
  • Massachusetts Institute of Technology (US)
  • ETH Zurich (CH)
Corresponding author
Kulik, Heather J.
ORCID
Venue
Industrial & Engineering Chemistry Research
Publisher
American Chemical Society (ACS)
Published
2018-09-24
Volume
57
Issue
42
Pages
13973-13986
ISSN
  • 0888-5885 (ISSN-L)
  • 1520-5045
DOI
10.1021/acs.iecr.8b04015
OpenAlex
W2892439819
Cited by
161
References
98
Citations by year
2026: 12 · 2025: 19 · 2024: 18 · 2023: 20 · 2022: 21 · 2021: 30 · 2020: 21 · 2019: 18 · 2018: 2
FWCI
6.63
Type
journal article
Language
en
Open access
Licence
cc-by-nc-nd
Locations
  • Industrial & Engineering Chemistry Research (publishedVersion)
  • ChemRxiv (acceptedVersion, cc-by-nc-nd)
  • OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
  • DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc)
  • Royal College of Surgeons in Ireland (submittedVersion, cc-by-nc-nd)
Funders
  • Defense Advanced Research Projects Agency (D18AP00039)
  • Massachusetts Institute of Technology (2014, 2017)
  • Office of Naval Research (N00014-17-1-2956, N00014-18-1-2434)
  • U.S. Department of Energy (DE-SC0018096)
  • Division of Chemical, Bioengineering, Environmental, and Transport Systems (CBET-1704266)
Topics
  • Physical Sciences › Materials Science › Materials Chemistry › Machine Learning in Materials Science
  • Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Advanced Chemical Physics Studies
  • Physical Sciences › Computer Science › Computational Theory and Mathematics › Computational Drug Discovery Methods
Keywords
machine learning, electronic structure prediction, HOMO energy level, HOMO-LUMO gap, descriptor selection, artificial neural networks, kernel ridge regression, LASSO, frontier molecular orbitals, steric effects, density functional theory

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review