QSAR modeling of the antimicrobial activity of peptides as a mathematical function of a sequence of amino acids

Explainable deep learning and virtual evolution identifies antimicrobial peptides with activity against multidrug-resistant human pathogens

Artificial intelligence (AI) is a promising approach to identify new antimicrobial compounds in diverse microbial species. Here we developed an AI-based, explainable deep learning model, EvoGradient, that predicts the potency of antimicrobial peptides (AMPs) and virtually modifies peptide sequences to produce more potent AMPs, akin to in silico directed evolution. We applied this model to peptides encoded in low-abundance human oral bacteria, resulting in the virtual evolution of 32 peptides into potent AMPs. Of these, the 6 most effective were synthesized and tested against multidrug-resistant pathogens and demonstrated activity against carbapenem-resistant species Escherichia coli, Klebsiella pneumoniae and Acinetobacter baumannii, and vancomycin-resistant Enterococcus faecium. The most potent AMP, pep-19-mod, was validated in vivo, achieving over 95% reduction in bacterial loads in mouse models of thigh infection through both systemic and local administration. Our approach advances the automatic identification and optimization of AMPs.

Aggregation-prone antimicrobial peptides target gram-negative bacterial nucleic acids and protein synthesis

Aggregation of antimicrobial peptides (AMPs) enhances their efficacy by destabilising the bacterial cell wall, membrane, and cytosolic proteins. Developing aggregation-prone AMPs offers a promising strategy to combat antibiotic resistance, though predicting such AMPs and understanding bacterial responses remain challenging. Octopus bimaculoides, a cephalopod species, lacks known AMP gene families, yet its protein fragments were used to predict AMPs via artificial intelligence tools. Four peptides (Oct-P1, Oct-P2, Oct-P3, and Oct-P4) were identified based on their aggregation propensity. Among them, Oct-P2 reduced the viability of Escherichia coli and Staphylococcus aureus by up to 90 %, confirmed by confocal laser scanning microscopy and scanning electron microscopy. It further aggregated plasmid DNA in vitro, and the presence of extracellular DNA reduced their antibacterial activity. With knockout mutants, it revealed that Oct-P2 was internalized into bacterial cells, possibly through membrane transport proteins, enhancing its antibacterial effect. Aggregation-induced emission assays and molecular dynamics simulations revealed that Oct-P2 aggregates with transcription promoter DNA, inhibiting transcription and translation in vitro. This dual-target mechanism not only highlights the potential of Oct-P2 as a lead template for new antimicrobial drug development, but also opens a new window for discovering AMPs from protein fragments against the upcoming challenge of bacterial infections. STATEMENT OF SIGNIFICANCE: A popular strategy for identifying antimicrobial peptides (AMPs) in specific genomes uses the conserved regions of AMP families, but this strategy has limitations in organisms lacking classical AMP gene families, such as Octopus. Fragments from non-antimicrobial proteins serve as a rich source for the identification of new AMPs. In this study, we used artificial intelligence tools to search for potential candidate AMP sequences from non-antimicrobial proteins in Octopus bimaculoides. The successful identification of aggregation-prone AMPs was shown to decrease bacterial viability, increase permeability, and reduce biomass. One candidate, Oct-P2, kills the gram-negative bacteria E. coli by aggregating with DNA and inhibiting transcription and translation, suggesting a new intracellular mechanism of AMP activity.

QSAR Reveals Decreased Lipophilicity of Polar Residues Determines the Selectivity of Antimicrobial Peptide Activity

Antimicrobial resistance has increased rapidly, causing daunting morbidity and mortality rates worldwide. Antimicrobial peptides (AMPs) have emerged as promising alternatives to traditional antibiotics due to their broad range of targets and low tendency to elicit resistance. However, potent antimicrobial activity is often accompanied by excessive cytotoxicity toward host cells, leading to a halt in AMP therapeutic development. Here, we present multivariate analyses that correlate 28 peptide properties to the activity and toxicity of 46 diverse African-derived AMPs and identify the negative lipophilicity of polar residues as an essential physiochemical property for selective antimicrobial activity. Twenty-seven active AMPs are identified, of which the majority are of scorpion or frog origin. Of these, thirteen are novel with no previously reported activities. Principal component analysis and quantitative structure-activity relationships (QSAR) reveal that overall hydrophobicity, lipophilicity, and residue side chain surface area affect the antimicrobial and cytotoxic activity of an AMP. This has been well documented previously, but the present QSAR analysis additionally reveals that a decrease in the lipophilicity, contributed by those amino acids classified as polar, confers selectivity for a peptide to pathogen over mammalian cells. Furthermore, an increase in overall peptide charge aids selectivity toward Gram-negative bacteria and fungi, while selectivity toward Gram-positive bacteria is obtained through an increased number of small lipophilic residues. Finally, a conservative increase in peptide size in terms of sequence length and molecular weight also contributes to improved activity without affecting toxicity. Our findings suggest a novel approach for the rational design or modification of existing AMPs to increase pathogen selectivity and enhance therapeutic potential.

Bioinformatics and bioactive peptides from foods: Do they work together?

We live in the Big Data Era which affects many aspects of science, including research on bioactive peptides derived from foods, which during the last few decades have been a focus of interest for scientists. These two issues, i.e., the development of computer technologies and progress in the discovery of novel peptides with health-beneficial properties, are closely interrelated. This Chapter presents the example applications of bioinformatics for studying biopeptides, focusing on main aspects of peptide analysis as the starting point, including: (i) the role of peptide databases; (ii) aspects of bioactivity prediction; (iii) simulation of peptide release from proteins. Bioinformatics can also be used for predicting other features of peptides, including ADMET, QSAR, structure, and taste. To answer the question asked "bioinformatics and bioactive peptides from foods: do they work together?", currently it is almost impossible to find examples of peptide research with no bioinformatics involved. However, theoretical predictions are not equivalent to experimental work and always require critical scrutiny. The aspects of compatibility of in silico and in vitro results are also summarized herein.

Identification of peptide epitopes of the gp120 protein of HIV-1 capable of inducing cellular and humoral immunity

The Human Immunodeficiency Virus (HIV-1) causes Acquired Immunodeficiency Syndrome (AIDS) and a high percentage of deaths. Therefore, it is necessary to design vaccines against HIV-1 for the prevention of AIDS. Bioinformatic tools and theoretical algorisms allow us to understand the structural proteins of viruses to develop vaccines based on immunogenic peptides (epitopes). In this work, we identified the epitopes: P1, P2, P10, P27 and P30 from the gp120 protein of HIV-1. These peptides were administered intranasally alone or with cholera toxin (CT) to BALB/c mice. The population of CD4+, CD8+ T lymphocytes and B cells (CD19/CD138+, IgA+ and IgG+) from nasal-associated lymphoid tissue, nasal passages, cervical and inguinal nodes was determined by flow cytometry. In addition, anti-peptides IgG and IgA from serum, nasal and vaginal washings were measured by ELISA. The results show that peptides administered by i.n. can modulate the immune response of T and B lymphocyte populations, as well as IgA and IgG antibodies secretion in the different sites analyzed. In conclusion, bioinformatics tools help us to select peptides with physicochemical properties that allow the induction of the humoral and cellular responses that depend on the peptide sequence.

A Quantitative Structure-Activity Relationship Approach to Determine Biotoxicity of Amide Herbicides for Ecotoxicological Risk Assessment

Amide herbicides have been widely applied in agriculture and found to be widespread and affect nontarget organisms in the environment. To better understand the biotoxicity mechanisms and determine the toxicity to the nontarget organisms for the hazard and risk assessment, five QSAR models were developed for the biotoxicity prediction of amide herbicides toward five aquatic and terrestrial organisms (including algae, daphnia, fish, earthworm and avian species), based on toxicity concentration and quantitative molecular descriptors. The results showed that the developed models complied with OECD principles for QSAR validation and presented excellent performances in predictive ability. In combination, the investigated QSAR relationship led to the toxicity mechanisms that eleven electrical descriptors (E_HOMO, E_LUMO, α_xx, α_yy, α_zz, μ, qN^-, Q_xx, Q_yy, qH⁺, and q^-), four thermodynamic descriptors (Cv, S^θ, H^θ, and ZPVE), and one steric descriptor (V_m) were strongly associated with the biotoxicity of amide herbicides. Electrical descriptors showed the greatest impacts on the toxicity of amide herbicides, followed by thermodynamic and steric descriptors.

The index of ideality of correlation improves the predictive potential of models of the antioxidant activity of tripeptides from frog skin (Litoria rubella)

It is usually held that good-quality models for the biological activity of peptides must take into account their 3D architecture and descriptors of quantum mechanics. However, the present study shows that it is possible to build up models without these complex calculations. The structure of tripeptides represented by sequences of one-symbol abbreviations of the corresponding amino acids serves to build up quantitative structure-activity relationships for the antioxidant activity of tripeptides from frog skin. The statistical quality of the best model for the validation set is n = 27, r² = 0.93, RMSE = 0.15.

The Monte Carlo Method as a Tool to Build up Predictive QSPR/QSAR

Background:

The Monte Carlo method has a wide application in various scientific researches. For the development of predictive models in a form of the quantitative structure-property / activity relationships (QSPRs/QSARs), the Monte Carlo approach also can be useful. The CORAL software provides the Monte Carlo calculations aimed to build up QSPR/QSAR models for different endpoints.

Methods:

Molecular descriptors are a mathematical function of so-called correlation weights of various molecular features. The numerical values of the correlation weights give the maximal value of a target function. The target function leads to a correlation between endpoint and optimal descriptor for the visible training set. The predictive potential of the model is estimated with the validation set, i.e. compounds that are not involved in the process of building up the model.

Results:

The approach gave quite good models for a large number of various physicochemical, biochemical, ecological, and medicinal endpoints. Bibliography and basic statistical characteristics of several CORAL models are collected in the present review. In addition, the extended version of the approach for more complex systems (nanomaterials and peptides), where behaviour of systems is defined by a group of conditions besides the molecular structure is demonstrated.

Conclusion:

The Monte Carlo technique available via the CORAL software can be a useful and convenient tool for the QSPR/QSAR analysis.

Computer-Aided Design of Antimicrobial Peptides: Are We Generating Effective Drug Candidates?

Antimicrobial peptides (AMPs), especially antibacterial peptides, have been widely investigated as potential alternatives to antibiotic-based therapies. Indeed, naturally occurring and synthetic AMPs have shown promising results against a series of clinically relevant bacteria. Even so, this class of antimicrobials has continuously failed clinical trials at some point, highlighting the importance of AMP optimization. In this context, the computer-aided design of AMPs has put together crucial information on chemical parameters and bioactivities in AMP sequences, thus providing modes of prediction to evaluate the antibacterial potential of a candidate sequence before synthesis. Quantitative structure-activity relationship (QSAR) computational models, for instance, have greatly contributed to AMP sequence optimization aimed at improved biological activities. In addition to machine-learning methods, the de novo design, linguistic model, pattern insertion methods, and genetic algorithms, have shown the potential to boost the automated design of AMPs. However, how successful have these approaches been in generating effective antibacterial drug candidates? Bearing this in mind, this review will focus on the main computational strategies that have generated AMPs with promising activities against pathogenic bacteria, as well as anti-infective potential in different animal models, including sepsis and cutaneous infections. Moreover, we will point out recent studies on the computer-aided design of antibiofilm peptides. As expected from automated design strategies, diverse candidate sequences with different structural arrangements have been generated and deposited in databases. We will, therefore, also discuss the structural diversity that has been engendered.

De Novo Design and In Vitro Testing of Antimicrobial Peptides against Gram-Negative Bacteria

Antimicrobial peptides (AMPs) have been identified as a potentially new class of antibiotics to combat bacterial resistance to conventional drugs. The design of de novo AMPs with high therapeutic indexes, low cost of synthesis, high resistance to proteases and high bioavailability remains a challenge. Such design requires computational modeling of antimicrobial properties. Currently, most computational methods cannot accurately calculate antimicrobial potency against particular strains of bacterial pathogens. We developed a tool for AMP prediction (Special Prediction (SP) tool) and made it available on our Web site (https://dbaasp.org/prediction). Based on this tool, a simple algorithm for the design of de novo AMPs (DSP) was created. We used DSP to design short peptides with high therapeutic indexes against gram-negative bacteria. The predicted peptides have been synthesized and tested in vitro against a panel of gram-negative bacteria, including drug resistant ones. Predicted activity against Escherichia coli ATCC 25922 was experimentally confirmed for 14 out of 15 peptides. Further improvements for designed peptides included the synthesis of D-enantiomers, which are traditionally used to increase resistance against proteases. One synthetic D-peptide (SP15D) possesses one of the lowest values of minimum inhibitory concentration (MIC) among all DBAASP database short peptides at the time of the submission of this article, while being highly stable against proteases and having a high therapeutic index. The mode of anti-bacterial action, assessed by fluorescence microscopy, shows that SP15D acts similarly to cell penetrating peptides. SP15D can be considered a promising candidate for the development of peptide antibiotics. We plan further exploratory studies with the SP tool, aiming at finding peptides which are active against other pathogenic organisms.

Heterologous Machine Learning for the Identification of Antimicrobial Activity in Human-Targeted Drugs

The emergence of microbes resistant to common antibiotics represent a current treat to human health. It has been recently recognized that non-antibiotic labeled drugs may promote antibiotic-resistance mechanisms in the human microbiome by presenting a secondary antibiotic activity; hence, the development of computer-assisted procedures to identify antibiotic activity in human-targeted compounds may assist in preventing the emergence of resistant microbes. In this regard, it is worth noting that while most antibiotics used to treat human infectious diseases are non-peptidic compounds, most known antimicrobials nowadays are peptides, therefore all computer-based models aimed to predict antimicrobials either use small datasets of non-peptidic compounds rendering predictions with poor reliability or they predict antimicrobial peptides that are not currently used in humans. Here we report a machine-learning-based approach trained to identify gut antimicrobial compounds; a unique aspect of our model is the use of heterologous training sets, in which peptide and non-peptide antimicrobial compounds were used to increase the size of the training data set. Our results show that combining peptide and non-peptide antimicrobial compounds rendered the best classification of gut antimicrobial compounds. Furthermore, this classification model was tested on the latest human-approved drugs expecting to identify antibiotics with broad-spectrum activity and our results show that the model rendered predictions consistent with current knowledge about broad-spectrum antibiotics. Therefore, heterologous machine learning rendered an efficient computational approach to classify antimicrobial compounds.

Unifying structural signature of eukaryotic α-helical host defense peptides

Diversity of α-helical host defense peptides (αHDPs) contributes to immunity against a broad spectrum of pathogens via multiple functions. Thus, resolving common structure-function relationships among αHDPs is inherently difficult, even for artificial-intelligence-based methods that seek multifactorial trends rather than foundational principles. Here, bioinformatic and pattern recognition methods were applied to identify a unifying signature of eukaryotic αHDPs derived from amino acid sequence, biochemical, and three-dimensional properties of known αHDPs. The signature formula contains a helical domain of 12 residues with a mean hydrophobic moment of 0.50 and favoring aliphatic over aromatic hydrophobes in 18-aa windows of peptides or proteins matching its semantic definition. The holistic α-core signature subsumes existing physicochemical properties of αHDPs, and converged strongly with predictions of an independent machine-learning-based classifier recognizing sequences inducing negative Gaussian curvature in target membranes. Queries using the α-core formula identified 93% of all annotated αHDPs in proteomic databases and retrieved all major αHDP families. Synthesis and antimicrobial assays confirmed efficacies of predicted sequences having no previously known antimicrobial activity. The unifying α-core signature establishes a foundational framework for discovering and understanding αHDPs encompassing diverse structural and mechanistic variations, and affords possibilities for deterministic design of antiinfectives.

Antimicrobial cell penetrating peptides with bacterial cell specificity: pharmacophore modelling, quantitative structure activity relationship and molecular dynamics simulation

Current research has shown cell-penetrating peptides and antimicrobial peptides (AMPs) as probable vectors for use in drug delivery and as novel antibiotics. It has been reported that the higher the therapeutic index (TI) the higher would be the bacterial cell penetrating ability. To the best of our knowledge, no in-silico study has been performed to determine bacterial cell specificity of the antimicrobial cell penetrating peptides (aCPP's) based on their TI. The aim of this study was to develop a quantitative structure activity relationship (QSAR) model, which can estimate antimicrobial potential and cell-penetrating ability of aCPPs against S. aureus, to confirm the relationship between the TI and aCPPs and to identify specific descriptors responsible for aCPPs penetrating ability. Molecular dynamics (MD) simulation was also performed to confirm the membrane insertion of the most active aCPPs obtained from the QSAR study. The most appropriate pharmacophore was identified to predict the aCPP's activity. The statistical results confirmed the validity of the model. The QSAR model was successful in identifying the optimal aCPP with high activity prediction and provided insights into the structural requirements to correlate their TI to cell penetrating ability. MD simulation of the best aCPP with 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) bilayer confirmed its interaction with the membrane and the C-terminal residues of the aCPP played a key role in membrane penetration. The strategy of combining QSAR and molecular dynamics, allowed for optimal estimation of ligand-target interaction and confirmed the importance of Trp and Lys in interacting with the POPC bilayer. Communicated by Ramaswamy H. Sarma.

Are the short cationic lipopeptides bacterial membrane disruptors? Structure-Activity Relationship and molecular dynamic evaluation

Short cationic lipopeptides are amphiphilic molecules that exhibit antimicrobial activity mainly against Gram-positives. These compounds bind to bacterial membranes and disrupt their integrity. Here we examine the structure-activity relation (SAR) of lysine-based lipopeptides, with a prospect to rationally design more active compounds. The presented study aims to explain how antimicrobial activity of lipopeptides is affected by the charge of lipopeptide headgroup and the length of lipopeptide acyl chain. The obtained SAR models suggest that the lipophilicity of short synthetic cationic lipopeptides is the major factor that determines their antimicrobial activities. In order to link the differences in antimicrobial activity to the mechanism of action of lipopeptides containing one and two hydrophobic chains, we additionally performed molecular dynamic (MD) simulations. By using combined coarse-grained and all-atom simulations we also show that these compounds neither affect the organization of the membrane lipids nor aggregate to form separate phases. These results, along with the onset of antimicrobial activity of lipopeptides well below the critical micelle concentration (CMC), indicate that lipopeptides do not act in a simple detergent-like manner.

The cationic tetradecapeptide mastoparan as a privileged structure for drug discovery: Enhanced antimicrobial properties of mitoparan analogues modified at position-14

Mastoparan (MP) peptides, distributed in insect venoms, induce a local inflammatory response post envenomation. Most endogenous MPs share common structural elements within a tetradecapeptide sequence that adopts an amphipathic helix whilst traversing biological membranes and when bound to an intracellular protein target. Rational modifications to increase cationic charge density and amphipathic helicity engineered mitoparan (MitP), a mitochondriotoxic bioportide and potent secretagogue. Following intracellular translocation, MitP is accreted by mitochondria thus indicating additional utility as an antimicrobial agent. Hence, the objectives of this study were to compare the antimicrobial activities of a structurally diverse set of cationic cell penetrating peptides, including both MP and MitP sequences, and to chemically engineer analogues of MitP for potential therapeutic applications. Herein, we confirm that, like MP, MitP is a privileged structure for the development of antimicrobial peptides active against both prokaryotic and eukaryotic pathogens. Collectively, MitP and target-selective chimeric analogues are broad spectrum antibiotics, with the Gram-negative A. baumannii demonstrating particular susceptibility. Modifications of MitP by amino acid substitution at position-14 produced peptides, Δ14MitP analogues, with unique pharmacodynamic properties. One example, [Ser¹⁴]MitP, lacks both cytotoxicity against human cell lines and mast cell secretory activity yet retains selective activity against the encapsulated yeast C. neoformans.

Recent trends and analytical challenges in plant bioactive peptide separation, identification and validation

Interest in research into bioactive peptides (BPs) is growing because of their health-promoting ability. Several bioactivities have been ascribed to peptides, including antioxidant, antihypertensive and antimicrobial properties. As they can be produced from precursor proteins, the investigation of BPs in foods is becoming increasingly popular. For the same reason, production of BPs from by-products has also emerged as a possible means of reducing waste and recovering value-added compounds suitable for functional food production and supplements. Milk, meat and fish are the most investigated sources of BPs, but vegetable-derived peptides are also of interest. Vegetables are commonly consumed, and agro-industrial wastes constitute a cheap, large and lower environmental impact source of proteins. The use of advanced analytical techniques for separation and identification of peptides would greatly benefit the discovery of new BPs. In this context, this review provides an overview of the most recent applications in BP investigations for vegetable food and by-products. The most important issues regarding peptide isolation and separation, by single or multiple chromatographic techniques, are discussed. Additionally, problems connected with peptide identification in plants and non-model plants are discussed regarding the particular case of BP identification. Finally, the issue of peptide validation to confirm sequence and bioactivity is presented. Graphical representation of the analytical workflow needed for investigation of bioactive peptides and applied to vegetables and vegetable wastes Graphical Abstract.

Rational Design of Membrane-Pore-Forming Peptides

Specific interactions of peptides with lipid membranes are essential for cellular communication and constitute a central aspect of the innate host defense against pathogens. A computational method for generating innovative membrane-pore-forming peptides inspired by natural templates is presented. Peptide representation in terms of sequence- and topology-dependent hydrophobic moments is introduced. This design concept proves to be appropriate for the de novo generation of first-in-class membrane-active peptides with the anticipated mode of action. The designed peptides outperform the natural template in terms of their antibacterial activity. They form a kinked helical structure and self-assemble in the membrane by an entropy-driven mechanism to form dynamically growing pores that are dependent on the lipid composition. The results of this study demonstrate the unique potential of natural template-based peptide design for chemical biology and medicinal chemistry.

Topological pattern for the search of new active drugs against methicillin resistant Staphylococcus aureus

Molecular topology was used to develop a mathematical model capable of classifying compounds according to antimicrobial activity against methicillin resistant Staphylococcus aureus (MRSA). Topological indices were used as structural descriptors and their relation to antimicrobial activity was determined by using linear discriminant analysis. This topological model establishes new structure activity relationships which show that the presence of cyclopropyl, chlorine and ramification pairs at a distance of two bonds favor this activity, while the presence of tertiary amines decreases it. This model was applied to a combinatorial library of a thousand and one 6-fluoroquinolones, from which 117 theoretical active molecules were obtained. The compound 10 and five new quinolones were tested against MRSA. They all showed some activity against MRSA, although compounds 6, 8 and 9 showed anti-MRSA activity similar to ciprofloxacin. This model was also applied to 263 theoretical antibacterial agents described by us in a previous work, from which 34 were predicted as theoretically active. Anti-MRSA activity was found bibliographically in 9 of them (ensuring at least 26% of success), and from the rest, 3 compounds were randomly chosen and tested, finding mitomycin C to be more active than ciprofloxacin. The results demonstrate the utility of the molecular topology approaches for identifying new drugs active against MRSA.

MD simulations and multivariate studies for modeling the antileishmanial activity of peptides

Leishmaniasis, a protozoan-caused disease, requires alternative treatments with minimized side-effects and less prone to resistance development. Antimicrobial peptides represent a possible choice to be developed. We report on the prospection of structural parameters of 23 helical antimicrobial and leishmanicidal peptides as a tool for modeling and predicting the activity of new peptides. This investigation is based on molecular dynamic simulations (MD) in mimetic membrane environment, as most of these peptides share the feature of interacting with phospholipid bilayers. To overcome the lack of experimental data on peptides' structures, we started simulations from designed 100% α-helices. This procedure was validated through comparisons with NMR data and the determination of the structure of Decoralin-amide. From physicochemical features and MD results, descriptors were raised and statistically related to the minimum inhibitory concentration against Leishmania by the multivariate data analysis technique. This statistical procedure confirmed five descriptors combined by different loadings in five principal components. The leishmanicidal activity depends on peptides' charge, backbone solvation, volume, and solvent-accessible surface area. The generated model possesses good predictability (q²  = 0.715, r²  = 0.898) and is indicative for the most and the least active peptides. This is a novel theoretical path for structure-activity studies combining computational methods that identify and prioritize the promising peptide candidates.

Quantitative Structure Activity Relationship study of the Anti-Hepatitis Peptides employing Random Forests and Extra-trees regressors

Antimicrobial peptides are host defense peptides being viewed as replacement to broad-spectrum antibiotics due to varied advantages. Hepatitis is the commonest infectious disease of liver, affecting 500 million globally with reported adverse side effects in treatment therapy. Antimicrobial peptides active against hepatitis are called as anti-hepatitis peptides (AHP). In current work, we present Extratrees and Random Forests based Quantitative Structure Activity Relationship (QSAR) regression modeling using extracted sequence based descriptors for prediction of the anti-hepatitis activity. The Extra-trees regression model yielded a very high performance in terms coefficient of determination (R2) as 0.95 for test set and 0.7 for the independent dataset. We hypothesize that the developed model can further be used to identify potentially active anti-hepatitis peptides with a high level of reliability.

PepBio: predicting the bioactivity of host defense peptides

A large-scale QSAR study of host defense peptides sheds light on the origin of their bioactivities (antibacterial, anticancer, antiviral and antifungal).

Enabling the Discovery and Virtual Screening of Potent and Safe Antimicrobial Peptides. Simultaneous Prediction of Antibacterial Activity and Cytotoxicity

Antimicrobial peptides (AMPs) represent promising alternatives to fight against bacterial pathogens. However, cellular toxicity remains one of the main concerns in the early development of peptide-based drugs. This work introduces the first multitasking (mtk) computational model focused on performing simultaneous predictions of antibacterial activities, and cytotoxicities of peptides. The model was created from a data set containing 3592 cases, and it displayed accuracy higher than 96% for classifying/predicting peptides in both training and prediction (test) sets. The technique known as alanine scanning was computationally applied to illustrate the calculation of the quantitative contributions of the amino acids (in their respective positions of the sequence) to the biological effects of a defined peptide. A small library formed by 10 peptides was generated, where peptides were designed by considering the interpretations of the different descriptors in the mtk-computational model. All the peptides were predicted to exhibit high antibacterial activities against multiple bacterial strains, and low cytotoxicity against various cell types. The present mtk-computational model can be considered a very useful tool to support high throughput research for the discovery of potent and safe AMPs.

First Multitarget Chemo-Bioinformatic Model To Enable the Discovery of Antibacterial Peptides against Multiple Gram-Positive Pathogens

Antimicrobial peptides (AMPs) have emerged as promising therapeutic alternatives to fight against the diverse infections caused by different pathogenic microorganisms. In this context, theoretical approaches in bioinformatics have paved the way toward the creation of several in silico models capable of predicting antimicrobial activities of peptides. All current models have several significant handicaps, which prevent the efficient search for highly active AMPs. Here, we introduce the first multitarget (mt) chemo-bioinformatic model devoted to performing alignment-free prediction of antibacterial activity of peptides against multiple Gram-positive bacterial strains. The model was constructed from a data set containing 2488 cases of AMPs sequences assayed against at least 1 out of 50 Gram-positive bacterial strains. This mt-chemo-bioinformatic model displayed percentages of correct classification higher than 90.00% in both training and prediction (test) sets. For the first time, two computational approaches derived from basic concepts in genetics and molecular biology were applied, allowing the calculations of the relative contributions of any amino acid (in a defined position) to the antibacterial activity of an AMP and depending on the bacterial strain used in the biological assay. The present mt-chemo-bioinformatic model constitutes a powerful tool to enable the discovery of potent and versatile AMPs.