2019
DOI: 10.1039/c9sc02696g
|View full text |Cite
|
Sign up to set email alerts
|

Electron density learning of non-covalent systems

Abstract: Machine learning model of the electron densities for analyzing non-covalent interaction patterns in peptides.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

5
142
0

Year Published

2019
2019
2023
2023

Publication Types

Select...
6
3

Relationship

0
9

Authors

Journals

citations
Cited by 131 publications
(147 citation statements)
references
References 98 publications
5
142
0
Order By: Relevance
“…ò ρ (%) was found to be 1.26% for the entire test set. This is very close to the errors reported in other studies [2,39]. The accuracy of the prediction can be visually observed from figure 6 where that charge density in each class of structures contained in the test set are plotted.…”
supporting
confidence: 88%
See 1 more Smart Citation
“…ò ρ (%) was found to be 1.26% for the entire test set. This is very close to the errors reported in other studies [2,39]. The accuracy of the prediction can be visually observed from figure 6 where that charge density in each class of structures contained in the test set are plotted.…”
supporting
confidence: 88%
“…An accurate estimate of ρ(r) can provide insights concerning charge redistribution, bond formation etc, in molecular and materials systems. ρ(r) is also the starting point for a variety of electronic structure simulations aimed at calculating higher level electronic properties like electrostatic moments associated with molecules such as dipole and quadrupole moments, electrostatic potentials (recently demonstrated by Fabrizio et al [2]), electrostatic interaction energies etc. It can also be used to directly compute IR intensities [3] and identify binding sites in host-guest compounds [4][5][6][7].…”
Section: Introductionmentioning
confidence: 99%
“…While these latter functionals can already be considered members of the machine learning (ML) family, other modern ML approaches make use of algorithms such as artificial neural networks (ANN), kernel ridge regression (KRR) and gaussian process regression (GPR). Grifasi et al [12] have shown that the electron density for small hydrocarbons can be directly predicted from structural information and Fabrizio et al [13] have been able to extend this work to non-covalently bonded systems. Chandrasekaran et al [14] were able to achieve the same goal for solid-state systems by introducing a grid-based structure to electron density mapping using an ANN.…”
Section: Introductionmentioning
confidence: 96%
“…While these latter functionals can already be considered members of the machine learning (ML) family, other modern ML approaches make use of algorithms such as artificial neural networks (ANN), kernel ridge regression (KRR) and gaussian process regression (GPR). Grifasi et al [18] have shown that the electron density for small hydrocarbons can be directly predicted from structural information and Fabrizio et al [19] have been able to extend this work to non-covalently bonded systems. Chandrasekaran et al [20] were able to achieve the same goal for solid-state systems by introducing a grid-based structure to electron density mapping using an ANN.…”
mentioning
confidence: 96%