An Interpretable Multitask Framework BiLAT Enables Accurate Prediction of Cyclin-Dependent Protein Kinase Inhibitors
Xu Qian1, Xiaowen Dai1, Lin Luo1
1Laboratory of Molecular Design and Drug Discovery, School of Science, China Pharmaceutical University, 639 Longmian Avenue, Nanjing 211198, China.
Abstract:
The cyclin-dependent protein kinases (CDKs) are protein-serine/threonine kinases with crucial effects on the regulation of cell cycle and transcription. CDKs can be a hallmark of cancer since their excessive expression could lead to impaired cell proliferation. However, the selectivity profile of most developed CDK inhibitors is not enough, which have hindered the therapeutic use of CDK inhibitors. In this study, we propose a multitask deep learning framework called BiLAT based on SMILES representation for the prediction of the inhibitory activity of molecules on eight CDK subtypes (CDK1, 2, 4-9). The framework is mainly composed of an improved bidirectional long short-term memory module BiLSTM and the encode layer of the Transformer framework. Additionally, the data enhancement method of SMILES enumeration is applied to improve the performance of the model. Compared with baseline predictive models based on three conventional machine learning methods and two multitask deep learning algorithms, BiLAT achieves the best performance with the highest average AUC, ACC, F1-score, and MCC values of 0.938, 0.894, 0.911, and 0.715 for the test set. Moreover, we constructed a targeted external data set CDK-Dec for the CDK family, which mainly contains bait values screened by 3D similarity with active compounds. This dataset was utilized in the subsequent evaluation of our model. It is worth mentioning that the BiLAT model is interpretable and can be used by chemists to design and synthesize compounds with improved activity. To further verify the generalization ability of the multitask BiLAT model, we also conducted another evaluation on three public datasets (Tox21, ClinTox, and SIDER). Compared with several currently popular models, BiLAT shows the best performance on two datasets. These results indicate that BiLAT is an effective tool for accelerating drug discovery.
Insights
A new deep learning model, BiLAT, accurately predicts cyclin-dependent protein kinase (CDK) inhibitor activity. This tool aids chemists in designing more effective cancer drugs by improving CDK inhibitor selectivity and accelerating drug discovery.
Area of Science:
- Biochemistry
- Computational Biology
- Drug Discovery
Background:
- Cyclin-dependent protein kinases (CDKs) regulate cell cycle and transcription, and their dysregulation is linked to cancer.
- Existing CDK inhibitors lack sufficient selectivity, hindering their therapeutic application.
- Developing selective CDK inhibitors is crucial for effective cancer treatment.
Purpose of the Study:
- To develop a multitask deep learning framework, BiLAT, for predicting molecular inhibitory activity against eight CDK subtypes.
- To enhance model performance using SMILES enumeration and evaluate its generalizability on public datasets.
Main Methods:
- A novel multitask deep learning framework, BiLAT, utilizing a bidirectional long short-term memory (BiLSTM) module and Transformer encoder.
- SMILES enumeration for data augmentation.
- Evaluation against conventional machine learning and other deep learning models on CDK and public datasets (Tox21, ClinTox, SIDER).
Main Results:
- BiLAT demonstrated superior performance compared to baseline models, achieving high average AUC (0.938), ACC (0.894), F1-score (0.911), and MCC (0.715) on the test set.
- The model showed strong performance on external and public datasets, indicating good generalization ability.
- BiLAT proved to be an interpretable tool for medicinal chemists.
Conclusions:
- BiLAT is an effective and interpretable deep learning tool for predicting CDK inhibitor activity.
- The framework accelerates the design and synthesis of novel compounds with improved therapeutic potential.
- BiLAT shows promise in advancing drug discovery for CDK-targeted therapies.
Related Concept Videos
02:34Inhibition of Cdk Activity
02:39Positive Regulator Molecules
02:15M-Cdk Drives Transition Into Mitosis
Cyclin-dependent kinases, or Cdks, work in concert with cyclins to control cell cycle transitions. M-Cdk, a complex of Cdk1 bound to M cyclin, is a well-known example of this coordinated control that drives the transition from the G2 to the M phase.
M cyclin...
01:28The Cell Cycle Control System
Cyclins and cyclin-dependent kinases (Cdks) are the primary cell cycle regulators and...
00:50Anaphase Promoting Complex
01:23Negative Regulator Molecules