POI-3DGCN: Predicting odor intensity of monomer flavors based on three-dimensionally embedded graph convolutional network

Liu, Qi; Luo, Dehan; Wen, Tengteng; GholamHosseini, Hamid; Qiu, Xinghui; Li, Jingshan

doi:10.1016/j.eswa.2022.116997

Cited by 8 publications

(3 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They update node information by exchanging information between linked nodes through message passing, and after multiple propagation steps, pooling operations are employed to extract molecule-level, global features. [1][2][3][4][5][6][7][8] Notable examples of c-GNN applications include antibiotic discovery, [9][10][11] retrosynthetic analysis, [12,13] and molecular dynamics simulation. [14][15][16] Conventional c-GNNs typically operate on 2D graphs, overlooking the valuable 3D information present in molecules.…”

Section: Introductionmentioning

confidence: 99%

“…c‐GNNs take molecular graphs as input, in which nodes are atoms, and edges are bonds in a molecule. They update node information by exchanging information between linked nodes through message passing, and after multiple propagation steps, pooling operations are employed to extract molecule‐level, global features [1–8] . Notable examples of c‐GNN applications include antibiotic discovery, [9–11] retrosynthetic analysis, [12,13] and molecular dynamics simulation [14–16] …”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MolNet_Equi: A Chemically Intuitive, Rotation‐Equivariant Graph Neural Network

Kim,

Jeong,

Kim

et al. 2023

Chemistry An Asian Journal

View full text Add to dashboard Cite

Although deep‐learning (DL) models suggest unprecedented prediction capabilities in tackling various chemical problems, their demonstrated tasks have so far been limited to the scalar properties including the magnitude of vectorial properties, such as molecular dipole moments. A rotation‐equivariant MolNet_Equi model, proposed in this paper, understands and recognizes the molecular rotation in the 3D Euclidean space, and exhibits the ability to predict directional dipole moments in the rotation‐sensitive mode, as well as showing superior performance for the prediction of scalar properties. Three consecutive operations of molecular rotation [[EQUATION]], dipole‐moment prediction [[EQUATION]], and dipole‐moment inverse‐rotation [[EQUATION]] do not alter the original prediction of the total dipole moment of a molecule [[EQUATION]], assuring the rotational equivariance of MolNet_Equi. Furthermore, MolNet_Equi faithfully predicts the absolute direction of dipole moments given molecular poses, albeit the model has been trained only with the information on dipole‐moment magnitudes, not directions. This work highlights the potential of incorporating fundamental yet crucial chemical rules and concepts into DL models, leading to the development of chemically intuitive models.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

MolNet_Equi: A Chemically Intuitive, Rotation‐Equivariant Graph Neural Network

Kim,

Jeong,

Kim

et al. 2023

Chemistry An Asian Journal

View full text Add to dashboard Cite

show abstract

“…During recent years, many researches have been conducted to predict odor perception of odorants using vairous parameters, such as electronic or physicochemical characteristics [9][10][11] , mass spectrometry (MS) 12 , and social network interactions 13 . Additionally, novel methods, such as odor-based social networks [14][15][16] , machine learning (ML) [17][18][19] , deep neural network (DNN) models [20][21][22] , and semantic-based approaches 23 have been developed to calibrate models that express the relationships between odorants and ODs. These studies have demonstrated the possibility to use data-driven approaches to solve the SOR problems.…”

Section: Introductionmentioning

confidence: 99%

Artificial intelligence-based gas chromatography-olfactometry for sensory evaluation of key compounds in food ingredients

Shang

Liu

Tang

et al. 2022

Preprint

View full text Add to dashboard Cite

The prediction of structure-odor-relationship (SOR) by deep neural networks (DNN) via the structural features of odorants has attracted great attention during the past decade. Due to the limited knowledge on binding mechanism between odorant molecules and olfactory receptors, however, it is not sure what kind of structural features play the most important role in smell recognition. Here, diverse DNNs, including molecular parameters neural network, molecular graphic convolution neural network (MG-CNN), molecular graph transformer neural network, and atom interaction neural network, were used to extract the structure features of molecules and to predict the categorized odor perception. The experimental results demonstrated that an MG-CNN combined with a multi-label DNN classifier produced the best results, with an area under the receiver operating characteristic curve and F1-score of 0.877±0.028 and 0.726±0.028, respectively. This is the first systematic study for molecular structure features extracted by different deep neural network and their predictive effect for SOR. We believe that these insights regarding the use of DNN-based odorant molecular feature extraction for odor sensory identification will be useful for introducing biologically interpretable artificial intelligence into olfactometry, and thus contribute to our understanding of the mechanisms underlying human olfaction.

show abstract

Utilizing deep learning to explore chemical space for drug lead optimization

Chakraborty

Hasija

2023

Expert Systems with Applications

View full text Add to dashboard Cite

POI-3DGCN: Predicting odor intensity of monomer flavors based on three-dimensionally embedded graph convolutional network

Cited by 8 publications

References 29 publications

MolNet_Equi: A Chemically Intuitive, Rotation‐Equivariant Graph Neural Network

MolNet_Equi: A Chemically Intuitive, Rotation‐Equivariant Graph Neural Network

Artificial intelligence-based gas chromatography-olfactometry for sensory evaluation of key compounds in food ingredients

Utilizing deep learning to explore chemical space for drug lead optimization

Contact Info

Product

Resources

About