MolHF: Molecular Heterogeneous Attributes Fusion for Drug-Target Affinity Prediction on Heterogeneity

Runze WANG; Zehua ZHANG; Yueqin ZHANG; Zhongyuan JIANG; Shilin SUN; Guixiang MA

doi:10.1587/transinf.2022DLP0023

MolHF: Molecular Heterogeneous Attributes Fusion for Drug-Target Affinity Prediction on Heterogeneity

Runze WANG, Zehua ZHANG, Yueqin ZHANG, Zhongyuan JIANG, Shilin SUN, Guixiang MA

Full Text Views

2

Cite this

Summary :

Recent studies in protein structure prediction such as AlphaFold have enabled deep learning to achieve great attention on the Drug-Target Affinity (DTA) task. Most works are dedicated to embed single molecular property and homogeneous information, ignoring the diverse heterogeneous information gains that are contained in the molecules and interactions. Motivated by this, we propose an end-to-end deep learning framework to perform Molecular Heterogeneous features Fusion (MolHF) for DTA prediction on heterogeneity. To address the challenges that biochemical attributes locates in different heterogeneous spaces, we design a Molecular Heterogeneous Information Learning module with multi-strategy learning. Especially, Molecular Heterogeneous Attention Fusion module is present to obtain the gains of molecular heterogeneous features. With these, the diversity of molecular structure information for drugs can be extracted. Extensive experiments on two benchmark datasets show that our method outperforms the baselines in all four metrics. Ablation studies validate the effect of attentive fusion and multi-group of drug heterogeneous features. Visual presentations demonstrate the impact of protein embedding level and the model ability of fitting data. In summary, the diverse gains brought by heterogeneous information contribute to drug-target affinity prediction.

Publication: IEICE TRANSACTIONS on Information Vol.E106-D No.5 pp.697-706

Publication Date: 2023/05/01

Publicized: 2022/05/31

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2022DLP0023

Type of Manuscript: Special Section PAPER (Special Section on Deep Learning Technologies: Architecture, Optimization, Techniques, and Applications)

Category: Smart Healthcare

Authors

Runze WANG
  Taiyuan University of Technology
Zehua ZHANG
  Taiyuan University of Technology
Yueqin ZHANG
  Taiyuan University of Technology
Zhongyuan JIANG
  Xidian University
Shilin SUN
  Taiyuan University of Technology
Guixiang MA
  University of Illinois at Chicago

Keyword

DTA prediction on heterogeneity, molecular heterogeneous information learning, molecular heterogeneous attention fusion

Cite this

Copy

Runze WANG, Zehua ZHANG, Yueqin ZHANG, Zhongyuan JIANG, Shilin SUN, Guixiang MA, "MolHF: Molecular Heterogeneous Attributes Fusion for Drug-Target Affinity Prediction on Heterogeneity" in IEICE TRANSACTIONS on Information, vol. E106-D, no. 5, pp. 697-706, May 2023, doi: 10.1587/transinf.2022DLP0023.
Abstract: Recent studies in protein structure prediction such as AlphaFold have enabled deep learning to achieve great attention on the Drug-Target Affinity (DTA) task. Most works are dedicated to embed single molecular property and homogeneous information, ignoring the diverse heterogeneous information gains that are contained in the molecules and interactions. Motivated by this, we propose an end-to-end deep learning framework to perform Molecular Heterogeneous features Fusion (MolHF) for DTA prediction on heterogeneity. To address the challenges that biochemical attributes locates in different heterogeneous spaces, we design a Molecular Heterogeneous Information Learning module with multi-strategy learning. Especially, Molecular Heterogeneous Attention Fusion module is present to obtain the gains of molecular heterogeneous features. With these, the diversity of molecular structure information for drugs can be extracted. Extensive experiments on two benchmark datasets show that our method outperforms the baselines in all four metrics. Ablation studies validate the effect of attentive fusion and multi-group of drug heterogeneous features. Visual presentations demonstrate the impact of protein embedding level and the model ability of fitting data. In summary, the diverse gains brought by heterogeneous information contribute to drug-target affinity prediction.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022DLP0023/_p

Copy

@ARTICLE{e106-d_5_697,
author={Runze WANG, Zehua ZHANG, Yueqin ZHANG, Zhongyuan JIANG, Shilin SUN, Guixiang MA, },
journal={IEICE TRANSACTIONS on Information},
title={MolHF: Molecular Heterogeneous Attributes Fusion for Drug-Target Affinity Prediction on Heterogeneity},
year={2023},
volume={E106-D},
number={5},
pages={697-706},
abstract={Recent studies in protein structure prediction such as AlphaFold have enabled deep learning to achieve great attention on the Drug-Target Affinity (DTA) task. Most works are dedicated to embed single molecular property and homogeneous information, ignoring the diverse heterogeneous information gains that are contained in the molecules and interactions. Motivated by this, we propose an end-to-end deep learning framework to perform Molecular Heterogeneous features Fusion (MolHF) for DTA prediction on heterogeneity. To address the challenges that biochemical attributes locates in different heterogeneous spaces, we design a Molecular Heterogeneous Information Learning module with multi-strategy learning. Especially, Molecular Heterogeneous Attention Fusion module is present to obtain the gains of molecular heterogeneous features. With these, the diversity of molecular structure information for drugs can be extracted. Extensive experiments on two benchmark datasets show that our method outperforms the baselines in all four metrics. Ablation studies validate the effect of attentive fusion and multi-group of drug heterogeneous features. Visual presentations demonstrate the impact of protein embedding level and the model ability of fitting data. In summary, the diverse gains brought by heterogeneous information contribute to drug-target affinity prediction.},
keywords={},
doi={10.1587/transinf.2022DLP0023},
ISSN={1745-1361},
month={May},}

Copy

TY - JOUR
TI - MolHF: Molecular Heterogeneous Attributes Fusion for Drug-Target Affinity Prediction on Heterogeneity
T2 - IEICE TRANSACTIONS on Information
SP - 697
EP - 706
AU - Runze WANG
AU - Zehua ZHANG
AU - Yueqin ZHANG
AU - Zhongyuan JIANG
AU - Shilin SUN
AU - Guixiang MA
PY - 2023
DO - 10.1587/transinf.2022DLP0023
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2023
AB - Recent studies in protein structure prediction such as AlphaFold have enabled deep learning to achieve great attention on the Drug-Target Affinity (DTA) task. Most works are dedicated to embed single molecular property and homogeneous information, ignoring the diverse heterogeneous information gains that are contained in the molecules and interactions. Motivated by this, we propose an end-to-end deep learning framework to perform Molecular Heterogeneous features Fusion (MolHF) for DTA prediction on heterogeneity. To address the challenges that biochemical attributes locates in different heterogeneous spaces, we design a Molecular Heterogeneous Information Learning module with multi-strategy learning. Especially, Molecular Heterogeneous Attention Fusion module is present to obtain the gains of molecular heterogeneous features. With these, the diversity of molecular structure information for drugs can be extracted. Extensive experiments on two benchmark datasets show that our method outperforms the baselines in all four metrics. Ablation studies validate the effect of attentive fusion and multi-group of drug heterogeneous features. Visual presentations demonstrate the impact of protein embedding level and the model ability of fitting data. In summary, the diverse gains brought by heterogeneous information contribute to drug-target affinity prediction.
ER -