Molecular substructure graph attention network for molecular property identification in drug discovery

Ye, Xian-bin; Guan, Quanlong; Luo, Weiqi; Fang, Liangda; Lai, Zhao-Rong; Wang, Jun

doi:10.1016/j.patcog.2022.108659

Cited by 11 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Self-supervised models includes Mol-CLR [36], GraphCL [37], HierMRL [38], GraphLoG [39], GraphMVP [32], GraphMAE [40], KEMPNN [41], MolPMoFiT [42], MolBERT [43], FP-BERT [27], and SMILES Transforme [15]. Supervised learning models includes D-MPNN [35], DimeNet [44], AttentionFP [45], DLF-MFF [46], and MSSGAT [47].…”

Section: Performance Comparison With Baselinesmentioning

confidence: 99%

Triple Generative Self-Supervised Learning Method for Molecular Property Prediction

Xu,

Xia,

Pan

et al. 2024

IJMS

View full text Add to dashboard Cite

Molecular property prediction is an important task in drug discovery, and with help of self-supervised learning methods, the performance of molecular property prediction could be improved by utilizing large-scale unlabeled dataset. In this paper, we propose a triple generative self-supervised learning method for molecular property prediction, called TGSS. Three encoders including a bi-directional long short-term memory recurrent neural network (BiLSTM), a Transformer, and a graph attention network (GAT) are used in pre-training the model using molecular sequence and graph structure data to extract molecular features. The variational auto encoder (VAE) is used for reconstructing features from the three models. In the downstream task, in order to balance the information between different molecular features, a feature fusion module is added to assign different weights to each feature. In addition, to improve the interpretability of the model, atomic similarity heat maps were introduced to demonstrate the effectiveness and rationality of molecular feature extraction. We demonstrate the accuracy of the proposed method on chemical and biological benchmark datasets by comparative experiments.

show abstract

Section: Performance Comparison With Baselinesmentioning

confidence: 99%

Triple Generative Self-Supervised Learning Method for Molecular Property Prediction

Xu,

Xia,

Pan

et al. 2024

IJMS

View full text Add to dashboard Cite

show abstract

“…Furthermore, the Schottky barrier (SB) in the source/drain was used in the design of TFETs to eliminate the doping effect of the S/D area, thus giving the TFETs better detection performance than the FETs. In addition, the introduction of high-k oxide [ 109 ] or charged plasma in TFETs [ 110 ] and the introduction of cavities on the source side for sensing biological molecules using dielectric modulation are helpful to enhance the drain current.…”

Section: Challenges and Opportunitiesmentioning

confidence: 99%

Application of Silicon Nanowire Field Effect Transistor (SiNW-FET) Biosensor with High Sensitivity

Li,

Chen

et al. 2023

Sensors

View full text Add to dashboard Cite

As a new type of one-dimensional semiconductor nanometer material, silicon nanowires (SiNWs) possess good application prospects in the field of biomedical sensing. SiNWs have excellent electronic properties for improving the detection sensitivity of biosensors. The combination of SiNWs and field effect transistors (FETs) formed one special biosensor with high sensitivity and target selectivity in real-time and label-free. Recently, SiNW-FETs have received more attention in fields of biomedical detection. Here, we give a critical review of the progress of SiNW-FETs, in particular, about the reversible surface modification methods. Moreover, we summarized the applications of SiNW-FETs in DNA, protein, and microbial detection. We also discuss the related working principle and technical approaches. Our review provides an extensive discussion for studying the challenges in the future development of SiNW-FETs.

show abstract

A multi-stream network for retrosynthesis prediction

Zhang,

Liu,

Zhang

et al. 2023

Front. Comput. Sci.

View full text Add to dashboard Cite

Molecular substructure graph attention network for molecular property identification in drug discovery

Cited by 11 publications

References 27 publications

Triple Generative Self-Supervised Learning Method for Molecular Property Prediction

Triple Generative Self-Supervised Learning Method for Molecular Property Prediction

Application of Silicon Nanowire Field Effect Transistor (SiNW-FET) Biosensor with High Sensitivity

A multi-stream network for retrosynthesis prediction

Contact Info

Product

Resources

About