FP-GNN: a versatile deep learning architecture for enhanced molecular property prediction

Cai, Hanxuan; Zhang, Huimin; Zhao, Duancheng; Wu, Jingxing

doi:10.1093/bib/bbac408

Cited by 61 publications

(49 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FP-GNN, a new DL architecture created in our Lab, can operate over a hybrid molecular representation of molecular graphs and fingerprints to enhance molecular property prediction ( Cai et al, 2022 ). However, there may be correlations between subtasks of datasets in the field of drug discovery and development, such as PARP subtypes.…”

Section: Resultsmentioning

confidence: 99%

“…FP-GNN finally uses fully convolutional networks (FCN) to fuse the features from both GNN and FPN, and then outputs the prediction results of molecular properties. FP-GNN DL algorithm achieves state-of-the-art (SOTA) performance on multiple molecular property prediction tasks ( Cai et al, 2022 ), and is freely available at ( https://github.com/idrugLab/FP-GNN ). However, most datasets in drug discovery have substantial connections between subtasks.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A multi-task FP-GNN framework enables accurate prediction of selective PARP inhibitors

Cai

et al. 2022

Front. Pharmacol.

Self Cite

View full text Add to dashboard Cite

PARP (poly ADP-ribose polymerase) family is a crucial DNA repair enzyme that responds to DNA damage, regulates apoptosis, and maintains genome stability; therefore, PARP inhibitors represent a promising therapeutic strategy for the treatment of various human diseases including COVID-19. In this study, a multi-task FP-GNN (Fingerprint and Graph Neural Networks) deep learning framework was proposed to predict the inhibitory activity of molecules against four PARP isoforms (PARP-1, PARP-2, PARP-5A, and PARP-5B). Compared with baseline predictive models based on four conventional machine learning methods such as RF, SVM, XGBoost, and LR as well as six deep learning algorithms such as DNN, Attentive FP, MPNN, GAT, GCN, and D-MPNN, the evaluation results indicate that the multi-task FP-GNN method achieves the best performance with the highest average BA, F1, and AUC values of 0.753 ± 0.033, 0.910 ± 0.045, and 0.888 ± 0.016 for the test set. In addition, Y-scrambling testing successfully verified that the model was not results of chance correlation. More importantly, the interpretability of the multi-task FP-GNN model enabled the identification of key structural fragments associated with the inhibition of each PARP isoform. To facilitate the use of the multi-task FP-GNN model in the field, an online webserver called PARPi-Predict and its local version software were created to predict whether compounds bear potential inhibitory activity against PARPs, thereby contributing to design and discover better selective PARP inhibitors.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

A multi-task FP-GNN framework enables accurate prediction of selective PARP inhibitors

Cai

et al. 2022

Front. Pharmacol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…FP‐GNN is a novel DL architecture for molecular property prediction, which combined and simultaneously studied the information from molecular graphs and molecular fingerprints 52–54 . The evaluation outcomes illustrated that FP‐GNN was a competitive DL algorithm for the molecular property prediction tasks.…”

Section: Methodsmentioning

confidence: 99%

“…FP-GNN is a novel DL architecture for molecular property prediction, which combined and simultaneously studied the information from molecular graphs and molecular fingerprints. [52][53][54] The evaluation outcomes illustrated that FP-GNN was a competitive DL algorithm for the molecular property prediction tasks. To achieve the best FP-GNN model, six hyperparameters are chosen to optimize by using the Hyperopt Pythonpackage 55 : the dropout rate of GAN, the number of multihead attentions, the hidden size of attentions, the hidden size and dropout rate of FPN, and the ratio of GNN in FP-GNN.…”

Section: Algorithms and Model Constructionmentioning

confidence: 99%

VDDB: A comprehensive resource and machine learning tool for antiviral drug discovery

Tao

Chen

et al. 2023

MedComm – Future Medicine

Self Cite

View full text Add to dashboard Cite

Virus infection is one of the major diseases that seriously threaten human health. To meet the growing demand for mining and sharing data resources related to antiviral drugs and to accelerate the design and discovery of new antiviral drugs, we presented an open‐access antiviral drug resource and machine learning platform (VDDB), which, to the best of our knowledge, is the first comprehensive dedicated resource for experimentally verified potential drugs/molecules based on manually curated data. Currently, VDDB highlights 848 clinical vaccines and 199 clinical antibodies, as well as over 710,000 small molecules targeting 39 medically important viruses including severe acute respiratory syndrome coronavirus 2. Furthermore, VDDB stores approximately three million records of pharmacological data for these collected potential antiviral drugs/molecules, involving 314 cell infection‐based phenotypic and 234 target‐based genotypic assays. Based on these annotated pharmacological data, VDDB allows users to browse, search, and download reliable information about these collects for various viruses of interest. In particular, VDDB also integrates 57 cell infection‐ and 117 target‐based associated high‐accuracy machine learning models to support various antivirals identification‐related tasks, such as compound activity prediction, virtual screening, drug repositioning, and target fishing. VDDB is freely accessible at https://vddb.idruglab.cn.

show abstract

“…Whether utilizing convolutional neural networks (CNNs) to process regular compound image information or leveraging recurrent neural networks (RNNs) and transformer architectures to operate on semantic molecular notations like SMILES, existing studies showed that these DL networks have led to tremendous advances in cheminformatics and computational chemistry, as well as drug design and discovery. Additionally, excellent works for predicting the properties of compounds by processing the raw graph information emerged more than two decades ago. − Recently, graph-based − molecular representation learning methods have shown comparable capabilities compared to other molecular representations (i.e., molecular descriptors and fingerprints) due to the superior capabilities of graph neural networks (GNNs) to model graph-structured data.…”

Section: Introductionmentioning

confidence: 99%

HiGNN: A Hierarchical Informative Graph Neural Network for Molecular Property Prediction Equipped with Feature-Wise Attention

Zhu

Zhang

Zhao

et al. 2022

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

Elucidating and accurately predicting the druggability and bioactivities of molecules plays a pivotal role in drug design and discovery and remains an open challenge. Recently, graph neural networks (GNNs) have made remarkable advancements in graph-based molecular property prediction. However, current graph-based deep learning methods neglect the hierarchical information of molecules and the relationships between feature channels. In this study, we propose a well-designed hierarchical informative graph neural network (termed HiGNN) framework for predicting molecular property by utilizing a corepresentation learning of molecular graphs and chemically synthesizable breaking of retrosynthetically interesting chemical substructure (BRICS) fragments. Furthermore, a plug-and-play feature-wise attention block is first designed in HiGNN architecture to adaptively recalibrate atomic features after the message passing phase. Extensive experiments demonstrate that HiGNN achieves state-of-the-art predictive performance on many challenging drug discovery-associated benchmark data sets. In addition, we devise a molecule-fragment similarity mechanism to comprehensively investigate the interpretability of the HiGNN model at the subgraph level, indicating that HiGNN as a powerful deep learning tool can help chemists and pharmacists identify the key components of molecules for designing better molecules with desired properties or functions. The source code is publicly available at .

show abstract

FP-GNN: a versatile deep learning architecture for enhanced molecular property prediction

Cited by 61 publications

References 55 publications

A multi-task FP-GNN framework enables accurate prediction of selective PARP inhibitors

A multi-task FP-GNN framework enables accurate prediction of selective PARP inhibitors

VDDB: A comprehensive resource and machine learning tool for antiviral drug discovery

HiGNN: A Hierarchical Informative Graph Neural Network for Molecular Property Prediction Equipped with Feature-Wise Attention

Contact Info

Product

Resources

About