loading page

Assessing Conformer Energies using Electronic Structure and Machine Learning Methods
  • Dakota Folmsbee,
  • Geoffrey Hutchison
Dakota Folmsbee
Department of Chemistry, University of Pittsburgh
Author Profile
Geoffrey Hutchison
Department of Chemistry, University of Pittsburgh

Corresponding Author:[email protected]

Author Profile


We have performed a large-scale evaluation of current computational methods, including conventional small-molecule force fields, semiempirical, density functional, ab initio electronic structure methods, and current machine learning (ML) techniques to evaluate relative single-point energies. Using up to 10 local minima geometries across ~700 molecules, each optimized by B3LYP-D3BJ with single-point DLPNO-CCSD(T) triple-zeta energies, we consider over 6,500 single points to compare the correlation between different methods for both relative energies and ordered rankings of minima. We find promise from current ML methods and recommend methods at each tier of the accuracy-time tradeoff, particularly the recent GFN2 semiempirical method, the B97-3c density functional approximation, and RI-MP2 for accurate conformer energies. The ANI family of ML methods shows promise, particularly the ANI-1ccx variant trained in part on coupled-cluster energies. Multiple methods suggest continued improvements should be expected in both performance and accuracy.
02 Mar 2020Submitted to IJQC Interactive Papers
11 Mar 2020Reviewer(s) Assigned
13 Apr 2020Review(s) Completed, Editorial Evaluation Pending
14 May 20201st Revision Received
12 Jun 2020Editorial Decision: Accept
05 Jan 2021Published in International Journal of Quantum Chemistry volume 121 issue 1. 10.1002/qua.26381
Olexandr Isayev posted a review
Referee ReportPaper by Folmsbee and Hutchison is a great example of reproducible benchmark paper and weel suited for IJQC special issue. Both codes and data are available on GitHub. This paper compares the accuracy of various computational methods to evaluate single point energies of molecular conformers. Authors used DLPNO-CCSD(T) as a reference level of theory and benchmarked small-molecule force fields, semiempirical, DFT and several emerging machine learning (ML) techniques. This paper provides computational chemists with a substantial body of high accuracy data. Overall this
Anonymous IJQC Reviewer posted a review
This is a follow-on paper from the Hutchison group, expanding on some previous work looking at correlations of molecular energy from a variety of levels of theory with results from high-level ab initio calculations. A new addition in this paper is a small set of ML methods, a welcome addition to the forcefield and electronic structure methods usually used in comparisons of this kind.The paper presents some interesting results, but is riddled with missing or misattributed data, typos, grammatical errors (particularly agreements for single
Geoffrey Hutchison and 1 more posted a review
Referee #1 (Report will be published at publication of the article)