ACP-MHCNN: an accurate multi-headed deep-convolutional neural network to predict anticancer peptides

Ahmed, Sajid; Muhammod, Rafsanjani; Khan, Zahid; Adilina, Sheikh; Sharma, Alok; Shatabda, Swakkhar; Dehzangi, Abdollah

doi:10.1038/s41598-021-02703-3

Cited by 49 publications

(33 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to verify whether the ensemble deep learning strategy of PD-BertEDL method can effectively solve the heterogeneity problem among different information and improve the prediction performance of peptide detectability, we designed and implemented three methods: Linear Fusion [ 26 ], Same Network [ 27 ], and Hybrid [ 38 , 39 , 40 ], respectively. Among them, Linear Fusion linearly spliced the three kinds of peptide information used in this paper, input the spliced vector into CNN+BiLSTM network to extract features and used softmax to classify.…”

Section: Resultsmentioning

confidence: 99%

PD-BertEDL: An Ensemble Deep Learning Method Using BERT and Multivariate Representation to Predict Peptide Detectability

Wang

Feng

et al. 2022

IJMS

View full text Add to dashboard Cite

Peptide detectability is defined as the probability of identifying a peptide from a mixture of standard samples, which is a key step in protein identification and analysis. Exploring effective methods for predicting peptide detectability is helpful for disease treatment and clinical research. However, most existing computational methods for predicting peptide detectability rely on a single information. With the increasing complexity of feature representation, it is necessary to explore the influence of multivariate information on peptide detectability. Thus, we propose an ensemble deep learning method, PD-BertEDL. Bidirectional encoder representations from transformers (BERT) is introduced to capture the context information of peptides. Context information, sequence information, and physicochemical information of peptides were combined to construct the multivariate feature space of peptides. We use different deep learning methods to capture the high-quality features of different categories of peptides information and use the average fusion strategy to integrate three model prediction results to solve the heterogeneity problem and to enhance the robustness and adaptability of the model. The experimental results show that PD-BertEDL is superior to the existing prediction methods, which can effectively predict peptide detectability and provide strong support for protein identification and quantitative analysis, as well as disease treatment.

show abstract

Section: Resultsmentioning

confidence: 99%

PD-BertEDL: An Ensemble Deep Learning Method Using BERT and Multivariate Representation to Predict Peptide Detectability

Wang

Feng

et al. 2022

IJMS

View full text Add to dashboard Cite

show abstract

“…In addition, these feature extraction methods can only extract the local order of peptide sequence through features such as DPC 23 , and it is difficult to grasp the global order information. Finally, the performance of these methods is largely related to manual feature extraction mechanisms, but it is not easy to extract suitable features for different data 26 .…”

Section: Introductionmentioning

confidence: 99%

“…Yi et al 28 proposed ACP-DL in 2019, which uses BPF and k-mer sparse matrix feature to represent the peptide sequences, and uses Long Short-Term Memory Model (LSTM) for prediction. Unlike traditional machine learning-based methods, deep learning-based methods do not require manual feature extraction to represent the input data 26 , that is, they can automatically extract features 29 . The methods based on deep learning can be divided into two categories: one uses deep learning methods to extract features, and then inputs the features into traditional machine learning classifiers such as SVM and RF for prediction; the other directly uses the deep learning method to make the final prediction.…”

Section: Introductionmentioning

confidence: 99%

“…In the second category, how the sequences are represented numerically is crucial 26 . Wu et al 31 adopted the word2vec word embedding method to encode sequences.…”

Section: Introductionmentioning

confidence: 99%

“…After the embedding layer, they used the encoder structure in the transformer 34 to predict ACPs, which requires many hyperparameters to be adjusted, including the number of self-attention heads, the dimensions of query, key, and value ,etc. In 2021, Ahmed et al 26 developed ACP-MHCNN based on a multi-head convolutional neural network, which uses BPF, physicochemical properties of amino acids, sequence evolution information to numerically represent peptide sequences. This numerical representation method integrates multiple aspects of information and has biological significance.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

ACPred-BMF: bidirectional LSTM with multiple feature representations for explainable anticancer peptide prediction

Han

Zhao

Zeng

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Cancer has become a major factor threatening human life and health. Under the circumstance that traditional treatment methods such as chemotherapy and radiotherapy are not highly specific and often cause severe side effects and toxicity, new treatment methods are urgently needed. Anticancer peptide drugs have low toxicity, stronger efficacy and specificity, and have emerged as a new type of cancer treatment drugs. However, experimental identification of anticancer peptides is time-consuming and expensive, and difficult to perform in a high-throughput manner. Computational identification of anticancer peptides can make up for the shortcomings of experimental identification. In this study, a deep learning-based predictor named ACPred-BMF is proposed for the prediction of anticancer peptides. This method uses the quantitative and qualitative properties of amino acids, binary profile feature to numerical representation for the peptide sequences. The Bidirectional LSTM network architecture is used in the model, and the attention mechanism is also considered. To alleviate the black-box problem of deep learning model prediction, we visualized the automatically extracted features and used the Shapley additive explanations algorithm to determine the importance of features to further understand the anticancer peptide mechanism. The results show that our method is one of the state-of-the-art anticancer peptide predictors. A web server as the implementation of ACPred-BMF that can be accessed via: http://mialab.ruc.edu.cn/ACPredBMFServer/.

show abstract

Efficient Mining of Anticancer Peptides from Gut Metagenome

Liu

Zhang

et al. 2023

Advanced Science

View full text Add to dashboard Cite

The gut microbiome plays a crucial role in modulating host health and disease. It serves as a vast reservoir of functional molecules that hold great potential for clinical applications. One specific area of interest is identifying anticancer peptides (ACPs) for innovative cancer therapies. However, ACPs discovery is hindered by a heavy reliance on experimental methodologies. To overcome this limitation, we here employed a novel approach by leveraging the overlap between ACPs and antimicrobial peptides (AMPs). By combining well‐established AMP prediction methods with mining techniques in metagenomic cohorts, a total of 40 potential ACPs is identified. Out of the identified ACPs, 39 demonstrated inhibitory effects against at least one cancer cell line, exhibiting significant differences from known ACPs. Moreover, the therapeutic potential of the two most promising peptides in a mouse xenograft cancer model is evaluated. Encouragingly, the peptides exhibit effective tumor inhibition without any detectable toxic effects. Interestingly, both peptides display uncommon secondary structures, highlighting its distinctive characteristics. This findings highlight the efficacy of the multi‐center mining approach, which effectively uncovers novel ACPs from the gut microbiome. This approach has significant implications for expanding treatment options not only for CRC, but also for other cancer types.

show abstract

ACP-MHCNN: an accurate multi-headed deep-convolutional neural network to predict anticancer peptides

Cited by 49 publications

References 53 publications

PD-BertEDL: An Ensemble Deep Learning Method Using BERT and Multivariate Representation to Predict Peptide Detectability

PD-BertEDL: An Ensemble Deep Learning Method Using BERT and Multivariate Representation to Predict Peptide Detectability

ACPred-BMF: bidirectional LSTM with multiple feature representations for explainable anticancer peptide prediction

Efficient Mining of Anticancer Peptides from Gut Metagenome

Contact Info

Product

Resources

About