Ensemble-AMPPred: Robust AMP Prediction and Recognition Using the Ensemble Learning Method with a New Hybrid Feature for Differentiating AMPs

Lertampaiporn, Supatcha; Vorapreeda, Tayvich; Hongsthong, Apiradee; Thammarongtham, Chinae

doi:10.3390/genes12020137

Cited by 22 publications

(17 citation statements)

References 58 publications

(29 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To select features that differentiate AMPs from non-AMPs, a feature selection method was almost used. Thus, the positive or negative samples (peptide sequences), according to their biological properties, are coded into numerical feature vectors that are used for learning the proposed model ( 135 ). Then, each peptide is encoded as a numerical feature vector based on suitable biological features, such as physicochemical properties, sequence composition, and structural features ( Figure 4b ).…”

Section: Recent Application Of Machine Learning Methods For Predictin...mentioning

confidence: 99%

“…In ensemble learning techniques, using multiple predictors and ensemble methods for incorporating individual classification models (bagging and boosting) leads to a decrease in FPs and increasing prediction accuracy. In this study, AMP prediction models were developed using ensemble methods based on five different algorithms, as well as comparing four different single models ( 135 ).…”

Section: Tools For Amps Predictionmentioning

confidence: 99%

“…Also, the similarity between the positive and negative datasets was removed, and as a result, the sequence similarity between the training and testing datasets was calculated as 47.29%. Lastly, the training data includes 1800 sequences of peptides from the AMP dataset, and 1800 sequences from the non-AMP dataset were prepared ( 135 ). A dataset consisting of 517 peptide features was then extracted and to filter this feature vector a logistic regression was applied to create a hybrid feature vector using four preselected single features based on an equation for logistic regression.…”

Section: Tools For Amps Predictionmentioning

confidence: 99%

See 2 more Smart Citations

A review on antimicrobial peptides databases and the computational tools

2022

View full text Add to dashboard Cite

Antimicrobial Peptides (AMPs) have been considered as potential alternatives for infection therapeutics since antibiotic resistance has been raised as a global problem. The AMPs are a group of natural peptides that play a crucial role in the immune system in various organisms AMPs have features such as a short length and efficiency against microbes. Importantly, they have represented low toxicity in mammals which makes them potential candidates for peptide-based drugs. Nevertheless, the discovery of AMPs is accompanied by several issues which are associated with labour-intensive and time-consuming wet-lab experiments. During the last decades, numerous studies have been conducted on the investigation of AMPs, either natural or synthetic type, and relevant data are recently available in many databases. Through the advancement of computational methods, a great number of AMP data are obtained from publicly accessible databanks, which are valuable resources for mining patterns to design new models for AMP prediction. However, due to the current flaws in assessing computational methods, more interrogations are warranted for accurate evaluation/analysis. Considering the diversity of AMPs and newly reported ones, an improvement in Machine Learning algorithms are crucial. In this review, we aim to provide valuable information about different types of AMPs, their mechanism of action and a landscape of current databases and computational tools as resources to collect AMPs and beneficial tools for the prediction and design of a computational model for new active AMPs.

show abstract

Section: Recent Application Of Machine Learning Methods For Predictin...mentioning

confidence: 99%

Section: Tools For Amps Predictionmentioning

confidence: 99%

Section: Tools For Amps Predictionmentioning

confidence: 99%

See 1 more Smart Citation

A review on antimicrobial peptides databases and the computational tools

2022

View full text Add to dashboard Cite

show abstract

“…Various composite features based on various combinations of informative selected features were built by using logistic regression based on the benchmarking data and then compared through a 10-fold cross-validation process. The detailed process of building composite features is described in the hybrid feature section of ensemble-AMPPred ( Lertampaiporn et al, 2021 ). A combination of features was used to fit a logistic regression model, which is represented by the following equation:

…”

Section: Methodsmentioning

confidence: 99%

Ensemble-AHTPpred: A Robust Ensemble Machine Learning Model Integrated With a New Composite Feature for Identifying Antihypertensive Peptides

et al. 2022

Self Cite

View full text Add to dashboard Cite

Hypertension or elevated blood pressure is a serious medical condition that significantly increases the risks of cardiovascular disease, heart disease, diabetes, stroke, kidney disease, and other health problems, that affect people worldwide. Thus, hypertension is one of the major global causes of premature death. Regarding the prevention and treatment of hypertension with no or few side effects, antihypertensive peptides (AHTPs) obtained from natural sources might be useful as nutraceuticals. Therefore, the search for alternative/novel AHTPs in food or natural sources has received much attention, as AHTPs may be functional agents for human health. AHTPs have been observed in diverse organisms, although many of them remain underinvestigated. The identification of peptides with antihypertensive activity in the laboratory is time- and resource-consuming. Alternatively, computational methods based on robust machine learning can identify or screen potential AHTP candidates prior to experimental verification. In this paper, we propose Ensemble-AHTPpred, an ensemble machine learning algorithm composed of a random forest (RF), a support vector machine (SVM), and extreme gradient boosting (XGB), with the aim of integrating diverse heterogeneous algorithms to enhance the robustness of the final predictive model. The selected feature set includes various computed features, such as various physicochemical properties, amino acid compositions (AACs), transitions, n-grams, and secondary structure-related information; these features are able to learn more information in terms of analyzing or explaining the characteristics of the predicted peptide. In addition, the tool is integrated with a newly proposed composite feature (generated based on a logistic regression function) that combines various feature aspects to enable improved AHTP characterization. Our tool, Ensemble-AHTPpred, achieved an overall accuracy above 90% on independent test data. Additionally, the approach was applied to novel experimentally validated AHTPs, obtained from recent studies, which did not overlap with the training and test datasets, and the tool could precisely predict these AHTPs.

show abstract

“…Research studies showed that DP7, an AMP designed in silico, showed broad-spectrum antimicrobial activity against MDR bacteria, such as P. aeruginosa [102]. Currently, there are many antimicrobial peptides databases (APDs) such as APD3 [63] and collection of antimicrobial peptides (CAMP)R3 [103], as well as online tools for AMP screening and identification such as dbAMP [104] and Ensemble-AMPPred [105].…”

Section: In Silico Designmentioning

confidence: 99%

Antimicrobial Peptides: From Design to Clinical Application

Zhang

Yang

2022

Antibiotics

View full text Add to dashboard Cite

Infection of multidrug-resistant (MDR) bacteria, such as methicillin-resistant Staphylococcus aureus (MRSA), carbapenem-resistant Enterobacteriaceae (CRE), and extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli, brings public health issues and causes economic burden. Pathogenic bacteria develop several methods to resist antibiotic killing or inhibition, such as mutation of antibiotic function sites, activation of drug efflux pumps, and enzyme-mediated drug degradation. Antibiotic resistance components can be transferred between bacteria by mobile genetic elements including plasmids, transposons, and integrons, as well as bacteriophages. The development of antibiotic resistance limits the treatment options for bacterial infection, especially for MDR bacteria. Therefore, novel or alternative antibacterial agents are urgently needed. Antimicrobial peptides (AMPs) display multiple killing mechanisms against bacterial infections, including directly bactericidal activity and immunomodulatory function, as potential alternatives to antibiotics. In this review, the development of antibiotic resistance, the killing mechanisms of AMPs, and especially, the design, optimization, and delivery of AMPs are reviewed. Strategies such as structural change, amino acid substitution, conjugation with cell-penetration peptide, terminal acetylation and amidation, and encapsulation with nanoparticles will improve the antimicrobial efficacy, reduce toxicity, and accomplish local delivery of AMPs. In addition, clinical trials in AMP studies or applications of AMPs within the last five years were summarized. Overall, AMPs display diverse mechanisms of action against infection of pathogenic bacteria, and future research studies and clinical investigations will accelerate AMP application.

show abstract

Ensemble-AMPPred: Robust AMP Prediction and Recognition Using the Ensemble Learning Method with a New Hybrid Feature for Differentiating AMPs

Cited by 22 publications

References 58 publications

A review on antimicrobial peptides databases and the computational tools

A review on antimicrobial peptides databases and the computational tools

Ensemble-AHTPpred: A Robust Ensemble Machine Learning Model Integrated With a New Composite Feature for Identifying Antihypertensive Peptides

Antimicrobial Peptides: From Design to Clinical Application

Contact Info

Product

Resources

About