Abstract:Abstract Motivation Anticancer peptides (ACPs) have recently emerged as effective anticancer drugs in cancer therapy. Machine learning-based predictors have been developed to identify ACPs and achieve satisfactory performance. However, existing methods suffer from experience-based feature engineering, which not only restricts the representation ability of the models to a certain extent but also lacks adaptivity for different data, limiting the further improvement of the predictive performance and impacting the robustness of the predictive models. To alleviate the above problems, we propose a novel deep-learning-based predictor named ACPred-LAF, in which we propose a novel multisense and multiscaled embedding algorithm to automatically learn and extract context sequential characteristics of ACPs. Results Through the feature comparative analysis, we demonstrate that our learnable and self-adaptive embedding features are better than hand-crafted features in capturing discriminative information, which can effectively benefit the performance improvement for ACP prediction. In addition, benchmarking comparison results demonstrate that our ACPred-LAF outperforms the state-of-the-art methods both on existing benchmark datasets and our newly constructed dataset. Furthermore, we also prove and validate the robustness of the model via the data interference experiment. To avoid potential evaluation bias, here, we construct a new ACP benchmark dataset named ACP-Mixed by integrating existing datasets. We expect our newly constructed dataset to be a golden standard benchmark dataset in this field. To facilitate the use of our model, we develop a web server as the implementation of ACPred-LAF. Availability and implementation Our proposed ACPred-LAF, newly constructed benchmark dataset ACP-Mixed are open source collaborative initiatives available in the GitHub repository (https://github.com/TearsWaiting/ACPred-LAF). Besides, a webserver as the implementation of ACPred-LAF that can be accessed via: http://server.malab.cn/ACPred-LAF. Supplementary information Supplementary data are available at Bioinformatics online.

PREDAIP: Computational Prediction and Analysis for Anti-inflammatory Peptide Via a Hybrid Feature Selection Technique

Predicting Anti-inflammatory Peptides by Ensemble Machine Learning and Deep Learning

iAIPs: Identifying Anti-Inflammatory Peptides Using Random Forest

DeepAIP: Deep learning for anti-inflammatory peptide prediction using pre-trained protein language model features based on contextual self-attention network

Comprehensive Analysis of Computational Methods for Predicting Anti-inflammatory Peptides

A BERT-based approach for identifying anti-inflammatory peptides using sequence information

A deep learning model for anti-inflammatory peptides identification based on deep variational autoencoder and contrastive learning

Large-scale comparative review and assessment of computational methods for anti-cancer peptide identification

Improved prediction of anti-angiogenic peptides based on machine learning models and comprehensive features from peptide sequences

Effective identification and differential analysis of anticancer peptides

ACEiPP: A Deep Learning-Based Framework to Predict Angiotensin-Converting Enzyme (ACE)-Inhibitory Peptides Using High-Efficiency Amino Acid Descriptors

PredAPP: Predicting Anti-Parasitic Peptides with Undersampling and Ensemble Approaches

ITP-Pred: an interpretable method for predicting, therapeutic peptides with fused features low-dimension representation

AIPs-SnTCN: Predicting Anti-Inflammatory Peptides Using fastText and Transformer Encoder-Based Hybrid Word Embedding with Self-Normalized Temporal Convolutional Networks

Learning embedding features based on multisense-scaled attention architecture to improve the predictive performance of anticancer peptides

miPEPPred-FRL: A Novel Method for Predicting Plant MiRNA-Encoded Peptides Using Adaptive Feature Representation Learning

mACPpred 2.0: Stacked Deep Learning for Anticancer Peptide Prediction with Integrated Spatial and Probabilistic Feature Representations

ACP-ML: A sequence-based method for anticancer peptide prediction

ACP-DA: Improving the Prediction of Anticancer Peptides Using Data Augmentation

Pred-AHCP: Robust feature selection enabled Sequence-Specific Prediction of Anti-Hepatitis C Peptides via Machine Learning

Comprehensive assessment of machine learning-based methods for predicting antimicrobial peptides