Yunyun Liang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yunyun Liang has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Shengli Zhang
School of Mathematics and Statistics, Xidian University, Xi’an, People’s Republic of China
7 shared publications Shengli Zhang
School of Mathematics and Statistics, Xidian University, Xi'an, 710071, PR China
4 shared publicationsHuijuan Qiao
School of Mathematics and Statistics, Xidian University, Xi'an, 710071, PR China
2 shared publicationsMengyi Cao
School of Science, Xi'an Polytechnic University, Xi'an, 710048, People's Republic of China
2 shared publicationsSanyang Liu
School of Mathematics and Statistics, Xidian University, Xi'an 710071, PR China
2 shared publicationsChenxia Wang
School of Science
1 shared publication
Latest Publications
- 8 May 2026
- Published by Springer Nature in Journal of Computer-Aided Molecular Design
- Vol. 40 (1)
NeuroPred-GMC: a dual-branch deep learning architecture for neuropeptide prediction based on gated dilated convolutional network and multi-scale convolutional network
Neuropeptides are multifunctional signaling molecules in the nervous system. By modulating synaptic transmission and integrating physiological systems, they influence a broad range of functions from pain perception to emotional regulation. Predicting neuropeptides can rapidly expand the library of potential therapeutic targets, thereby providing novel candidate molecules for drug development in areas such as analgesics, anti-anxiety medications, and weight-loss drugs. Traditional experimental methods are extremely time-consuming, labor-intensive, these promising alternative computational methods have emerged. In this study, a dual-branch deep learning architecture for neuropeptide prediction known as NeuroPred-GMC are built up based on gated dilated convolutional network with ESM-2 feature representation and multi-scale convolutional network with Prot-T5 feature representation. Dilated convolution exponentially enlarges the receptive field via increased dilation rates, gating mechanism enables dynamic, selective feature enhancement and noise suppression, and multi-scale convolution captures multi-level contextual information. On the independence test set, the accuracy of 93.24%, Sn of 93.69%, Sp of 92.79%, Pre of 92.86%, MCC of 0.8649 and the auROC of 0.9667 are obtained. The experimental results through cross-validation and independent test demonstrate that the proposed model has good robustness and generalizability, and can serve as a supplemental candidate predictor. The source datasets and codes can be freely available at https://github.com/yunyunliang88/NeuroPred-GMC .
- 8 May 2026
- Published by Springer Nature in Journal of Computer-Aided Molecular Design
- Vol. 40 (1)
NeuroPred-GMC: a dual-branch deep learning architecture for neuropeptide prediction based on gated dilated convolutional network and multi-scale convolutional network
Neuropeptides are multifunctional signaling molecules in the nervous system. By modulating synaptic transmission and integrating physiological systems, they influence a broad range of functions from pain perception to emotional regulation. Predicting neuropeptides can rapidly expand the library of potential therapeutic targets, thereby providing novel candidate molecules for drug development in areas such as analgesics, anti-anxiety medications, and weight-loss drugs. Traditional experimental methods are extremely time-consuming, labor-intensive, these promising alternative computational methods have emerged. In this study, a dual-branch deep learning architecture for neuropeptide prediction known as NeuroPred-GMC are built up based on gated dilated convolutional network with ESM-2 feature representation and multi-scale convolutional network with Prot-T5 feature representation. Dilated convolution exponentially enlarges the receptive field via increased dilation rates, gating mechanism enables dynamic, selective feature enhancement and noise suppression, and multi-scale convolution captures multi-level contextual information. On the independence test set, the accuracy of 93.24%, Sn of 93.69%, Sp of 92.79%, Pre of 92.86%, MCC of 0.8649 and the auROC of 0.9667 are obtained. The experimental results through cross-validation and independent test demonstrate that the proposed model has good robustness and generalizability, and can serve as a supplemental candidate predictor. The source datasets and codes can be freely available at https://github.com/yunyunliang88/NeuroPred-GMC .
- 23 March 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (7) , 4058-4074
ToxPLTC: Peptide Toxicity Prediction by Integrating Pretrained T5 Protein Language Model and Text Convolutional Neural Network
Peptide-based therapeutics show promising potential in treating a range of diseases, such as diabetes, cancer, and chronic pain. However, critical challenges, including peptide toxicity, immunogenicity, and stability deficiencies of peptides, have become major obstacles to their direct clinical application. Traditional toxicity testing methods based on a wet lab are not only costly but also time-consuming. In contrast, the classification model based on deep learning provides a new technical path for the efficient identification of peptide toxicity. In this study, we propose a deep learning framework called ToxPLTC, which employs the ProtT5 protein language model based on the Transformer architecture for pretraining peptide sequences, adopts the borderline SMOTE algorithm to handle an imbalanced training set data, and utilizes a text convolutional neural network combined with a fully connected layer for classification. Additionally, visualization analysis, motif analysis, and mutation-scan analysis are performed to understand the function of each module and enhance the interpretability of our model. The applicability domain is constructed based on the K-NN strategy to define the effective prediction range of our model to ensure the reliability of model predictions. The ToxPLTC model achieves a balanced accuracy of 93.02% on independent test set 1 and 88.04% on independent test set 2. Experimental results demonstrate that our model outperforms existing models on independent test sets and has good generalization ability. The ToxPLTC model possesses great potential as a valuable and robust tool for peptide-based drug development. The source data sets and codes can be available at the following GitHub repository: https://github.com/yunyunliang88/ToxPLTC.
- 23 March 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (7) , 4058-4074
ToxPLTC: Peptide Toxicity Prediction by Integrating Pretrained T5 Protein Language Model and Text Convolutional Neural Network
Peptide-based therapeutics show promising potential in treating a range of diseases, such as diabetes, cancer, and chronic pain. However, critical challenges, including peptide toxicity, immunogenicity, and stability deficiencies of peptides, have become major obstacles to their direct clinical application. Traditional toxicity testing methods based on a wet lab are not only costly but also time-consuming. In contrast, the classification model based on deep learning provides a new technical path for the efficient identification of peptide toxicity. In this study, we propose a deep learning framework called ToxPLTC, which employs the ProtT5 protein language model based on the Transformer architecture for pretraining peptide sequences, adopts the borderline SMOTE algorithm to handle an imbalanced training set data, and utilizes a text convolutional neural network combined with a fully connected layer for classification. Additionally, visualization analysis, motif analysis, and mutation-scan analysis are performed to understand the function of each module and enhance the interpretability of our model. The applicability domain is constructed based on the K-NN strategy to define the effective prediction range of our model to ensure the reliability of model predictions. The ToxPLTC model achieves a balanced accuracy of 93.02% on independent test set 1 and 88.04% on independent test set 2. Experimental results demonstrate that our model outperforms existing models on independent test sets and has good generalization ability. The ToxPLTC model possesses great potential as a valuable and robust tool for peptide-based drug development. The source data sets and codes can be available at the following GitHub repository: https://github.com/yunyunliang88/ToxPLTC.
- Open Access
- 29 May 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-12
A deep learning model for prediction of lysine crotonylation sites by fusing multi-features based on multi-head self-attention mechanism
Lysine crotonylation (Kcr) is an important post-translational modification, which is present in both histone and non-histone proteins, and plays a key role in a variety of biological processes such as metabolism and cell differentiation. Therefore, rapid and accurate identification of this modification has become a key task to study its biological effects. In the past few years, some calculation methods have been developed, but there is room for improvement in prediction performance. In this paper, we propose an effective model named DeepMM-Kcr, which is based on multiple features and an innovative deep learning framework. Multiple features are extracted from natural language processing features and hand-crafted features, where natural language processing features include token embedding and positional embedding encoded by transformer, and hand-crafted features include one-hot, amino acid index and position-weighted amino acid composition, and encoded by bidirectional long short-term memory network. Then natural language processing features and hand-crafted features are fusing by multi-head self-attention mechanism. Finally, a deep learning framework is constructed based on convolutional neural network, bidirectional gated recurrent unit and multilayer perceptron for robust prediction of Kcr sites. On the independent test set, the accuracy of DeepMM-Kcr is highest among the existing models. The experimental results show that our model has very good performance in predicting Kcr sites. The source datasets and codes (in Python) are publicly available at https://github.com/yunyunliang88/DeepMM-Kcr.
- 29 May 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-12
Open AccessA deep learning model for prediction of lysine crotonylation sites by fusing multi-features based on multi-head self-attention mechanism
Lysine crotonylation (Kcr) is an important post-translational modification, which is present in both histone and non-histone proteins, and plays a key role in a variety of biological processes such as metabolism and cell differentiation. Therefore, rapid and accurate identification of this modification has become a key task to study its biological effects. In the past few years, some calculation methods have been developed, but there is room for improvement in prediction performance. In this paper, we propose an effective model named DeepMM-Kcr, which is based on multiple features and an innovative deep learning framework. Multiple features are extracted from natural language processing features and hand-crafted features, where natural language processing features include token embedding and positional embedding encoded by transformer, and hand-crafted features include one-hot, amino acid index and position-weighted amino acid composition, and encoded by bidirectional long short-term memory network. Then natural language processing features and hand-crafted features are fusing by multi-head self-attention mechanism. Finally, a deep learning framework is constructed based on convolutional neural network, bidirectional gated recurrent unit and multilayer perceptron for robust prediction of Kcr sites. On the independent test set, the accuracy of DeepMM-Kcr is highest among the existing models. The experimental results show that our model has very good performance in predicting Kcr sites. The source datasets and codes (in Python) are publicly available at https://github.com/yunyunliang88/DeepMM-Kcr.
- 16 August 2024
- Published by Elsevier in Analytical Biochemistry
- Vol. 695, 115648
NeuroPred-ResSE: Predicting neuropeptides by integrating residual block and squeeze-excitation attention mechanism
- 16 August 2024
- Published by Elsevier in Analytical Biochemistry
- Vol. 695, 115648
NeuroPred-ResSE: Predicting neuropeptides by integrating residual block and squeeze-excitation attention mechanism
- Open Access
- 1 January 2023
- Published by American Institute of Mathematical Sciences (AIMS) in Mathematical Biosciences and Engineering
- Vol. 20 (12) , 21563-21587
iAVPs-ResBi: Identifying antiviral peptides by using deep residual network and bidirectional gated recurrent unit
Human history is also the history of the fight against viral diseases. From the eradication of viruses to coexistence, advances in biomedicine have led to a more objective understanding of viruses and a corresponding increase in the tools and methods to combat them. More recently, antiviral peptides (AVPs) have been discovered, which due to their superior advantages, have achieved great impact as antiviral drugs. Therefore, it is very necessary to develop a prediction model to accurately identify AVPs. In this paper, we develop the iAVPs-ResBi model using k-spaced amino acid pairs (KSAAP), encoding based on grouped weight (EBGW), enhanced grouped amino acid composition (EGAAC) based on the N5C5 sequence, composition, transition and distribution (CTD) based on physicochemical properties for multi-feature extraction. Then we adopt bidirectional long short-term memory (BiLSTM) to fuse features for obtaining the most differentiated information from multiple original feature sets. Finally, the deep model is built by combining improved residual network and bidirectional gated recurrent unit (BiGRU) to perform classification. The results obtained are better than those of the existing methods, and the accuracies are 95.07, 98.07, 94.29 and 97.50% on the four datasets, which show that iAVPs-ResBi can be used as an effective tool for the identification of antiviral peptides. The datasets and codes are freely available at https://github.com/yunyunliang88/iAVPs-ResBi.
- 1 January 2023
- Published by American Institute of Mathematical Sciences (AIMS) in Mathematical Biosciences and Engineering
- Vol. 20 (12) , 21563-21587
Open AccessiAVPs-ResBi: Identifying antiviral peptides by using deep residual network and bidirectional gated recurrent unit
Human history is also the history of the fight against viral diseases. From the eradication of viruses to coexistence, advances in biomedicine have led to a more objective understanding of viruses and a corresponding increase in the tools and methods to combat them. More recently, antiviral peptides (AVPs) have been discovered, which due to their superior advantages, have achieved great impact as antiviral drugs. Therefore, it is very necessary to develop a prediction model to accurately identify AVPs. In this paper, we develop the iAVPs-ResBi model using k-spaced amino acid pairs (KSAAP), encoding based on grouped weight (EBGW), enhanced grouped amino acid composition (EGAAC) based on the N5C5 sequence, composition, transition and distribution (CTD) based on physicochemical properties for multi-feature extraction. Then we adopt bidirectional long short-term memory (BiLSTM) to fuse features for obtaining the most differentiated information from multiple original feature sets. Finally, the deep model is built by combining improved residual network and bidirectional gated recurrent unit (BiGRU) to perform classification. The results obtained are better than those of the existing methods, and the accuracies are 95.07, 98.07, 94.29 and 97.50% on the four datasets, which show that iAVPs-ResBi can be used as an effective tool for the identification of antiviral peptides. The datasets and codes are freely available at https://github.com/yunyunliang88/iAVPs-ResBi.
- 23 December 2022
- Published by Taylor & Francis in SAR and QSAR in Environmental Research
- Vol. 34 (1) , 1-19
iACP-GE: accurate identification of anticancer peptides by using gradient boosting decision tree and extra tree
Cancer is one of the main diseases threatening human life, accounting for millions of deaths around the world each year. Traditional physical and chemical methods for cancer treatment are extremely time-consuming, lab-intensive, expensive, inefficient and difficult to be applied in a high-throughput way. Hence, it is an urgent task to develop automated computational methods to enable fast and accurate identification of anticancer peptides (ACPs). In this paper, we develop a novel model named iACP-GE to identify ACPs. Multi-features are extracted by using binary encoding, enhanced grouped amino acid composition and BLOSUM62 encoding based on the N5C5 sequence, as well as detrended forward moving-average auto-cross correlation analysis based on physicochemical properties of 20 natural amino acids. Thus, 835 features are obtained for each sample, in order to avoid information redundancy, gradient boosting decision tree was adopted as the feature selection strategy. Then, the optimal feature subset is input to the extra tree classifier. The accuracies of ACP740 and ACP240 datasets with the 5-fold cross-validation were 90.54% and 91.25%, respectively. Experimental results indicate that iACP-GE significantly outperforms several existing models on ACP740 and ACP240 datasets and can be used as an effective tool for the identification of ACPs. The datasets and source codes for iACP-GE are available at https://github.com/yunyunliang88/iACP-GE.
- 23 December 2022
- Published by Taylor & Francis in SAR and QSAR in Environmental Research
- Vol. 34 (1) , 1-19
iACP-GE: accurate identification of anticancer peptides by using gradient boosting decision tree and extra tree
Cancer is one of the main diseases threatening human life, accounting for millions of deaths around the world each year. Traditional physical and chemical methods for cancer treatment are extremely time-consuming, lab-intensive, expensive, inefficient and difficult to be applied in a high-throughput way. Hence, it is an urgent task to develop automated computational methods to enable fast and accurate identification of anticancer peptides (ACPs). In this paper, we develop a novel model named iACP-GE to identify ACPs. Multi-features are extracted by using binary encoding, enhanced grouped amino acid composition and BLOSUM62 encoding based on the N5C5 sequence, as well as detrended forward moving-average auto-cross correlation analysis based on physicochemical properties of 20 natural amino acids. Thus, 835 features are obtained for each sample, in order to avoid information redundancy, gradient boosting decision tree was adopted as the feature selection strategy. Then, the optimal feature subset is input to the extra tree classifier. The accuracies of ACP740 and ACP240 datasets with the 5-fold cross-validation were 90.54% and 91.25%, respectively. Experimental results indicate that iACP-GE significantly outperforms several existing models on ACP740 and ACP240 datasets and can be used as an effective tool for the identification of ACPs. The datasets and source codes for iACP-GE are available at https://github.com/yunyunliang88/iACP-GE.
- 1 March 2022
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 32 (3) , 033131
Data driven adaptive Gaussian mixture model for solving Fokker–Planck equation
The Fokker–Planck (FP) equation provides a powerful tool for describing the state transition probability density function of complex dynamical systems governed by stochastic differential equations (SDEs). Unfortunately, the analytical solution of the FP equation can be found in very few special cases. Therefore, it has become an interest to find a numerical approximation method of the FP equation suitable for a wider range of nonlinear systems. In this paper, a machine learning method based on an adaptive Gaussian mixture model (AGMM) is proposed to deal with the general FP equations. Compared with previous numerical discretization methods, the proposed method seamlessly integrates data and mathematical models. The prior knowledge generated by the assumed mathematical model can improve the performance of the learning algorithm. Also, it yields more interpretability for machine learning methods. Numerical examples for one-dimensional and two-dimensional SDEs with one and/or two noises are given. The simulation results show the effectiveness and robustness of the AGMM technique for solving the FP equation. In addition, the computational complexity and the optimization algorithm of the model are also discussed.
- 1 March 2022
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 32 (3) , 033131
Data driven adaptive Gaussian mixture model for solving Fokker–Planck equation
The Fokker–Planck (FP) equation provides a powerful tool for describing the state transition probability density function of complex dynamical systems governed by stochastic differential equations (SDEs). Unfortunately, the analytical solution of the FP equation can be found in very few special cases. Therefore, it has become an interest to find a numerical approximation method of the FP equation suitable for a wider range of nonlinear systems. In this paper, a machine learning method based on an adaptive Gaussian mixture model (AGMM) is proposed to deal with the general FP equations. Compared with previous numerical discretization methods, the proposed method seamlessly integrates data and mathematical models. The prior knowledge generated by the assumed mathematical model can improve the performance of the learning algorithm. Also, it yields more interpretability for machine learning methods. Numerical examples for one-dimensional and two-dimensional SDEs with one and/or two noises are given. The simulation results show the effectiveness and robustness of the AGMM technique for solving the FP equation. In addition, the computational complexity and the optimization algorithm of the model are also discussed.
- 1 October 2021
- Published by Elsevier in Analytical Biochemistry
- Vol. 630, 114335
iPromoter-ET: Identifying promoters and their strength by extremely randomized trees-based feature selection
- 1 October 2021
- Published by Elsevier in Analytical Biochemistry
- Vol. 630, 114335
iPromoter-ET: Identifying promoters and their strength by extremely randomized trees-based feature selection
- 1 May 2021
- Published by Bentham Science Publishers Ltd. in Current Pharmaceutical Design
- Vol. 27 (17) , 2076-2087
Application of Machine Learning Techniques in Drug-target Interactions Prediction
Background: Drug-Target interactions are vital for drug design and drug repositioning. However, traditional lab experiments are both expensive and time-consuming. Various computational methods which applied machine learning techniques performed efficiently and effectively in the field. Results: The machine learning methods can be divided into three categories basically: Supervised methods, Semi-Supervised methods and Unsupervised methods. We reviewed recent representative methods applying machine learning techniques of each category in DTIs and summarized a brief list of databases frequently used in drug discovery. In addition, we compared the advantages and limitations of these methods in each category. Conclusion: Every prediction model has both strengths and weaknesses and should be adopted in proper ways. Three major problems in DTIs prediction including the lack of nonreactive drug-target pairs data sets, over optimistic results due to the biases and the exploiting of regression models on DTIs prediction should be seriously considered.
- 1 May 2021
- Published by Bentham Science Publishers Ltd. in Current Pharmaceutical Design
- Vol. 27 (17) , 2076-2087
Application of Machine Learning Techniques in Drug-target Interactions Prediction
Background: Drug-Target interactions are vital for drug design and drug repositioning. However, traditional lab experiments are both expensive and time-consuming. Various computational methods which applied machine learning techniques performed efficiently and effectively in the field. Results: The machine learning methods can be divided into three categories basically: Supervised methods, Semi-Supervised methods and Unsupervised methods. We reviewed recent representative methods applying machine learning techniques of each category in DTIs and summarized a brief list of databases frequently used in drug discovery. In addition, we compared the advantages and limitations of these methods in each category. Conclusion: Every prediction model has both strengths and weaknesses and should be adopted in proper ways. Three major problems in DTIs prediction including the lack of nonreactive drug-target pairs data sets, over optimistic results due to the biases and the exploiting of regression models on DTIs prediction should be seriously considered.
- Open Access
- 1 January 2021
- Published by American Institute of Mathematical Sciences (AIMS) in Mathematical Biosciences and Engineering
- Vol. 18 (6) , 8797-8814
iEnhancer-MFGBDT: Identifying enhancers and their strength by fusing multiple features and gradient boosting decision tree
Enhancer is a non-coding DNA fragment that can be bound with proteins to activate transcription of a gene, hence play an important role in regulating gene expression. Enhancer identification is very challenging and more complicated than other genetic factors due to their position variation and free scattering. In addition, it has been proved that genetic variation in enhancers is related to human diseases. Therefore, identification of enhancers and their strength has important biological meaning. In this paper, a novel model named iEnhancer-MFGBDT is developed to identify enhancer and their strength by fusing multiple features and gradient boosting decision tree (GBDT). Multiple features include k-mer and reverse complement k-mer nucleotide composition based on DNA sequence, and second-order moving average, normalized Moreau-Broto auto-cross correlation and Moran auto-cross correlation based on dinucleotide physical structural property matrix. Then we use GBDT to select features and perform classification successively. The accuracies reach 78.67% and 66.04% for identifying enhancers and their strength on the benchmark dataset, respectively. Compared with other models, the results show that our model is useful and effective intelligent tool to identify enhancers and their strength, of which the datasets and source codes are available at https://github.com/shengli0201/iEnhancer-MFGBDT1.
- 1 January 2021
- Published by American Institute of Mathematical Sciences (AIMS) in Mathematical Biosciences and Engineering
- Vol. 18 (6) , 8797-8814
Open AccessiEnhancer-MFGBDT: Identifying enhancers and their strength by fusing multiple features and gradient boosting decision tree
Enhancer is a non-coding DNA fragment that can be bound with proteins to activate transcription of a gene, hence play an important role in regulating gene expression. Enhancer identification is very challenging and more complicated than other genetic factors due to their position variation and free scattering. In addition, it has been proved that genetic variation in enhancers is related to human diseases. Therefore, identification of enhancers and their strength has important biological meaning. In this paper, a novel model named iEnhancer-MFGBDT is developed to identify enhancer and their strength by fusing multiple features and gradient boosting decision tree (GBDT). Multiple features include k-mer and reverse complement k-mer nucleotide composition based on DNA sequence, and second-order moving average, normalized Moreau-Broto auto-cross correlation and Moran auto-cross correlation based on dinucleotide physical structural property matrix. Then we use GBDT to select features and perform classification successively. The accuracies reach 78.67% and 66.04% for identifying enhancers and their strength on the benchmark dataset, respectively. Compared with other models, the results show that our model is useful and effective intelligent tool to identify enhancers and their strength, of which the datasets and source codes are available at https://github.com/shengli0201/iEnhancer-MFGBDT1.