A deep learning method for predicting the minimum inhibitory concentration of antimicrobial peptides against Escherichia coli using Multi-Branch-CNN and Attention.

Antimicrobial peptides (AMPs) are a promising alternative to antibiotics to combat drug resistance in pathogenic bacteria. However, the development of AMPs with high potency and specificity remains a challenge, and new tools to evaluate antimicrobial activity are needed to accelerate the discovery process. Therefore, we proposed MBC-Attention, a combination of a multi-branch convolution neural network architecture and attention mechanisms to predict the experimental minimum inhibitory concentration of peptides against Escherichia coli. The optimal MBC-Attention model achieved an average Pearson correlation coefficient (PCC) of 0.775 and a root mean squared error (RMSE) of 0.533 (log µM) in three independent tests of randomly drawn sequences from the data set. This results in a 5-12% improvement in PCC and a 6-13% improvement in RMSE compared to 17 traditional machine learning models and 2 optimally tuned models using random forest and support vector machine. Ablation studies confirmed that the two proposed attention mechanisms, global attention and local attention, contributed largely to performance improvement. IMPORTANCE Antimicrobial peptides (AMPs) are potential candidates for replacing conventional antibiotics to combat drug resistance in pathogenic bacteria. Therefore, it is necessary to evaluate the antimicrobial activity of AMPs quantitatively. However, wet-lab experiments are labor-intensive and time-consuming. To accelerate the evaluation process, we develop a deep learning method called MBC-Attention to regress the experimental minimum inhibitory concentration of AMPs against Escherichia coli. The proposed model outperforms traditional machine learning methods. Data, scripts to reproduce experiments, and the final production models are available on GitHub.

Discovery and analysis of a novel antimicrobial peptide B1AW from the skin secretion of Amolops wuyiensis and improving the membrane-binding affinity through the construction of the lysine-introduced analogue.

In the development and study of antimicrobial peptides (AMPs), researchers have kept a watchful eye on peptides from the brevinin family because of their extensive antimicrobial activities and anticancer potency. In this study, a novel brevinin peptide was isolated from the skin secretions of the Wuyi torrent frog, Amolops wuyiensis (A. wuyiensisi), named B1AW (FLPLLAGLAANFLPQIICKIARKC). B1AW displayed anti-bacterial activity against Gram-positive bacteria Staphylococcus aureus (S. aureus), methicillin-resistant Staphylococcus aureus (MRSA), and Enterococcus faecalis (E. faecalis). B1AW-K was designed to broaden the antimicrobial spectrum of B1AW. The introduction of a lysine residue generated an AMP with enhanced broad-spectrum antibacterial activity. It also displayed the ability to inhibit the growth of human prostatic cancer PC-3, non-small lung cancer H838, and glioblastoma cancer U251MG cell lines. In molecular dynamic (MD) simulations, B1AW-K had a faster approach and adsorption to the anionic membrane than B1AW. Therefore, B1AW-K was considered a drug prototype with a dual effect, which deserves further clinical investigation and validation.

Recent Progress in the Discovery and Design of Antimicrobial Peptides Using Traditional Machine Learning and Deep Learning.

Antimicrobial resistance has become a critical global health problem due to the abuse of conventional antibiotics and the rise of multi-drug-resistant microbes. Antimicrobial peptides (AMPs) are a group of natural peptides that show promise as next-generation antibiotics due to their low toxicity to the host, broad spectrum of biological activity, including antibacterial, antifungal, antiviral, and anti-parasitic activities, and great therapeutic potential, such as anticancer, anti-inflammatory, etc. Most importantly, AMPs kill bacteria by damaging cell membranes using multiple mechanisms of action rather than targeting a single molecule or pathway, making it difficult for bacterial drug resistance to develop. However, experimental approaches used to discover and design new AMPs are very expensive and time-consuming. In recent years, there has been considerable interest in using in silico methods, including traditional machine learning (ML) and deep learning (DL) approaches, to drug discovery. While there are a few papers summarizing computational AMP prediction methods, none of them focused on DL methods. In this review, we aim to survey the latest AMP prediction methods achieved by DL approaches. First, the biology background of AMP is introduced, then various feature encoding methods used to represent the features of peptide sequences are presented. We explain the most popular DL techniques and highlight the recent works based on them to classify AMPs and design novel peptide sequences. Finally, we discuss the limitations and challenges of AMP prediction.

Multi-Branch-CNN: Classification of ion channel interacting peptides using multi-branch convolutional neural network.

Ligand peptides that have high affinity for ion channels are critical for regulating ion flux across the plasma membrane. These peptides are now being considered as potential drug candidates for many diseases, such as cardiovascular disease and cancers. In this work, we developed Multi-Branch-CNN, a CNN method with multiple input branches for identifying three types of ion channel peptide binders (sodium, potassium, and calcium) from intra- and inter-feature types. As for its real-world applications, prediction models that are able to recognize novel sequences having high or low similarities to training sequences are required. To this end, we tested our models on two test sets: a general test set including sequences spanning different similarity levels to those of the training set, and a novel-test set consisting of only sequences that bear little resemblance to sequences from the training set. Our experiments showed that the Multi-Branch-CNN method performs better than thirteen traditional ML algorithms (TML13), yielding an improvement in accuracy of 3.2%, 1.2%, and 2.3% on the test sets as well as 8.8%, 14.3%, and 14.6% on the novel-test sets for sodium, potassium, and calcium ion channels, respectively. We confirmed the effectiveness of Multi-Branch-CNN by comparing it to the standard CNN method with one input branch (Single-Branch-CNN) and an ensemble method (TML13-Stack). The data sets, script files to reproduce the experiments, and the final predictive models are freely available at https://github.com/jieluyan/Multi-Branch-CNN.

Chemical toxicity prediction based on semi-supervised learning and graph convolutional neural network.

As safety is one of the most important properties of drugs, chemical toxicology prediction has received increasing attentions in the drug discovery research. Traditionally, researchers rely on in vitro and in vivo experiments to test the toxicity of chemical compounds. However, not only are these experiments time consuming and costly, but experiments that involve animal testing are increasingly subject to ethical concerns. While traditional machine learning (ML) methods have been used in the field with some success, the limited availability of annotated toxicity data is the major hurdle for further improving model performance. Inspired by the success of semi-supervised learning (SSL) algorithms, we propose a Graph Convolution Neural Network (GCN) to predict chemical toxicity and trained the network by the Mean Teacher (MT) SSL algorithm. Using the Tox21 data, our optimal SSL-GCN models for predicting the twelve toxicological endpoints achieve an average ROC-AUC score of 0.757 in the test set, which is a 6% improvement over GCN models trained by supervised learning and conventional ML methods. Our SSL-GCN models also exhibit superior performance when compared to models constructed using the built-in DeepChem ML methods. This study demonstrates that SSL can increase the prediction power of models by learning from unannotated data. The optimal unannotated to annotated data ratio ranges between 1:1 and 4:1. This study demonstrates the success of SSL in chemical toxicity prediction; the same technique is expected to be beneficial to other chemical property prediction tasks by utilizing existing large chemical databases. Our optimal model SSL-GCN is hosted on an online server accessible through: https://app.cbbio.online/ssl-gcn/home .

xDeep-AcPEP: Deep Learning Method for Anticancer Peptide Activity Prediction Based on Convolutional Neural Network and Multitask Learning.

Cancer is one of the leading causes of death worldwide. Conventional cancer treatment relies on radiotherapy and chemotherapy, but both methods bring severe side effects to patients, as these therapies not only attack cancer cells but also damage normal cells. Anticancer peptides (ACPs) are a promising alternative as therapeutic agents that are efficient and selective against tumor cells. Here, we propose a deep learning method based on convolutional neural networks to predict biological activity (EC50, LC50, IC50, and LD50) against six tumor cells, including breast, colon, cervix, lung, skin, and prostate. We show that models derived with multitask learning achieve better performance than conventional single-task models. In repeated 5-fold cross validation using the CancerPPD data set, the best models with the applicability domain defined obtain an average mean squared error of 0.1758, Pearson's correlation coefficient of 0.8086, and Kendall's correlation coefficient of 0.6156. As a step toward model interpretability, we infer the contribution of each residue in the sequence to the predicted activity by means of feature importance weights derived from the convolutional layers of the model. The present method, referred to as xDeep-AcPEP, will help to identify effective ACPs in rational peptide design for therapeutic purposes. The data, script files for reproducing the experiments, and the final prediction models can be downloaded from http://github.com/chen709847237/xDeep-AcPEP. The web server to directly access this prediction method is at https://app.cbbio.online/acpep/home.

A novel bioengineered fragment peptide of Vasostatin-1 exerts smooth muscle pharmacological activities and anti-angiogenic effects via blocking VEGFR signalling pathway.

Chromogranin A (CgA) is a hydrophilic glycoprotein released by post-ganglionic sympathetic neurons. CgA consists of a single peptide chain containing numerous paired basic residues, which are typical cleavage sites in prohormones to generate bioactive peptides. It is recognized as a diagnostic and prognostic serum marker for neuroendocrine tumours. Vasostatin-1 is one of the most conserved regions of CgA and has diverse inhibitory biological activities. In this study, a novel peptide fragment that contains three typical functional structures of Vasostatin-1 was synthesized. This unique bioengineered Vasostatin-1 Derived Peptide (named V1DP) includes a highly conserved domain between vertebrate species in its N-terminal region, comprising a disulphide bridge formed by two cysteine residues at amino acid positions 17 and 38, respectively. Besides, V1DP contains two significant tripeptide recognition sequences: the amino acid triplets, RGD and KGD. Our data demonstrated that V1DP could induce a dose-dependent relaxation of rat arterial smooth muscle and also increase the contraction activity of rat uterus smooth muscle. More importantly, we found that V1DP inhibits cancer cell proliferation, modulate the HUVEC cell migration, and exhibit anti-angiogenesis effect both in vitro and in vivo. We further investigated the actual mechanism of V1DP, and our results confirmed that V1DP involves inhibiting the vascular endothelial growth factor receptor (VEGFR) signalling. We docked V1DP to the apo structures of VEGFR2 and examined the stability of the peptide in the protein pockets. Our simulation and free energy calculations results indicated that V1DP can bind to the catalytic domain and regulatory domain pockets, depending on whether the conformational state of the protein is JM-in or JM-out. Taken together, our data suggested that V1DP plays a role as the regulator of endothelial cell function and smooth muscle pharmacological homeostasis. V1DP is a water-soluble and biologically stable peptide and could further develop as an anti-angiogenic drug for cancer treatment.

LigTMap: ligand and structure-based target identification and activity prediction for small molecular compounds.

Target prediction is a crucial step in modern drug discovery. However, existing experimental approaches to target prediction are time-consuming and costly. Here, we introduce LigTMap, an online server with a fully automated workflow that can identify protein targets of chemical compounds among 17 classes of therapeutic proteins extracted from the PDBbind database. It combines ligand similarity search with docking and binding similarity analysis to predict putative targets. In the validation experiment of 1251 compounds, targets were successfully predicted for more than 70% of the compounds within the top-10 list. The performance of LigTMap is comparable to the current best servers SwissTargetPrediction and SEA. When testing with our newly compiled compounds from recent literature, we get improved top 10 success rate (66% ours vs. 60% SwissTargetPrediction and 64% SEA) and similar top 1 success rate (45% ours vs. 51% SwissTargetPrediction and 41% SEA). LigTMap directly provides ligand docking structures in PDB format, so that the results are ready for further structural studies in computer-aided drug design and drug repurposing projects. The LigTMap web server is freely accessible at https://cbbio.online/LigTMap . The source code is released on GitHub ( https://github.com/ShirleyWISiu/LigTMap ) under the BSD 3-Clause License to encourage re-use and further developments.

Aggregation and Its Influence on the Bioactivities of a Novel Antimicrobial Peptide, Temporin-PF, and Its Analogues.

Temporin is an antimicrobial peptide (AMP) family discovered in the skin secretion of ranid frog that has become a promising alternative for conventional antibiotic therapy. Herein, a novel temporin peptide, Temporin-PF (TPF), was successfully identified from Pelophylax fukienensis. It exhibited potent activity against Gram-positive bacteria, but no effect on Gram-negative bacteria. Additionally, TPF exhibited aggregation effects in different solutions. Three analogs were further designed to study the relationship between the aggregation patterns and bioactivities, and the MD simulation was performed for revealing the pattern of the peptide assembly. As the results showed, all peptides were able to aggregate in the standard culture media and salt solutions, especially CaCl2 and MgCl2 buffers, where the aggregation was affected by the concentration of the salts. MD simulation reported that all peptides were able to form oligomers. The parent peptide assembly depended on the hydrophobic interaction via the residues in the middle domain of the sequence. However, the substitution of Trp/D-Trp resulted in an enhanced inter-peptide interaction in the zipper-like domain and eliminated overall biological activities. Our study suggested that introducing aromaticity at the zipper-like domain for temporin may not improve the bioactivities, which might be related to the formation of aggregates via the inter-peptide contacts at the zipper-like motif domain, and it could reduce the binding affinity to the lipid membrane of microorganisms.

Effect of Concentration, Chain Length, Hydrophobicity, and an External Electric Field on the Growth of Mixed Alkanethiol Self-Assembled Monolayers: A Molecular Dynamics Study.

Growing functionalized self-assembled monolayers (SAMs) with fewer defects and lower cost is the focus of ongoing investigations. In the present study, molecular dynamics simulations were performed to investigate the process of SAM formation on a gold substrate from mixed alkanethiolates in ethanol solution. Using the mixed-SAM system of 11-mercaptoundecanoic acid (MUA) with either 1-decanethiol (C9CH3) or 6-mercaptohexanol (C6OH) in a 3:7 ratio as the standard SAM model, we systematically investigated the effects of the concentration, chain length, functional group, and an external electric field on SAM growth. The results showed that the initial growth rate and surface coverage of the SAM are dependent on the ligand concentration. At a certain high concentration (about 1.2-1.5 times the minimum concentration), the final surface coverage is optimal. Reducing the chain length and increasing the proportion of hydrophobic diluting molecules are effective ways to improve the surface coverage, but the compositional ligands have to be changed, which may not be desirable for the functional requirements of SAMs. Furthermore, by investigating the behavior of the alkanethiolates and ethanol solvent under an applied external field, we find that a strong electric field with a proper field direction can facilitate the generation of defect-free monolayers. These findings will contribute to the understanding of mixed-SAM formation and provide insight into experimental design for efficient and effective SAM formation.

GWOVina: A grey wolf optimization approach to rigid and flexible receptor docking.

Protein-ligand docking programs are indispensable tools for predicting the binding pose of a ligand to the receptor protein. In this paper, we introduce an efficient flexible docking method, GWOVina, which is a variant of the Vina implementation using the grey wolf optimizer (GWO) and random walk for the global search, and the Dunbrack rotamer library for side-chain sampling. The new method was validated for rigid and flexible-receptor docking using four independent datasets. In rigid docking, GWOVina showed comparable docking performance to Vina in terms of ligand pose RMSD, success rate, and affinity prediction. In flexible-receptor docking, GWOVina has improved success rate compared to Vina and AutoDockFR. It ran 2 to 7 times faster than Vina and 40 to 100 times faster than AutoDockFR. Therefore, GWOVina can play a role in solving the complex flexible-receptor docking cases and is suitable for virtual screening of compound libraries. GWOVina is freely available at https://cbbio.cis.um.edu.mo/software/gwovina for testing.

White learning methodology: A case study of cancer-related disease factors analysis in real-time PACS environment.

BACKGROUND AND OBJECTIVE: Bayesian network is a probabilistic model of which the prediction accuracy may not be one of the highest in the machine learning family. Deep learning (DL) on the other hand possess of higher predictive power than many other models. How reliable the result is, how it is deduced, how interpretable the prediction by DL mean to users, remain obscure. DL functions like a black box. As a result, many medical practitioners are reductant to use deep learning as the only tool for critical machine learning application, such as aiding tool for cancer diagnosis. METHODS: In this paper, a framework of white learning is being proposed which takes advantages of both black box learning and white box learning. Usually, black box learning will give a high standard of accuracy and white box learning will provide an explainable direct acyclic graph. According to our design, there are 3 stages of White Learning, loosely coupled WL, semi coupled WL and tightly coupled WL based on degree of fusion of the white box learning and black box learning. In our design, a case of loosely coupled WL is tested on breast cancer dataset. This approach uses deep learning and an incremental version of Naïve Bayes network. White learning is largely defied as a systemic fusion of machine learning models which result in an explainable Bayes network which could find out the hidden relations between features and class and deep learning which would give a higher accuracy of prediction than other algorithms. We designed a series of experiments for this loosely coupled WL model. RESULTS: The simulation results show that using WL compared to standard black-box deep learning, the levels of accuracy and kappa statistics could be enhanced up to 50%. The performance of WL seems more stable too in extreme conditions such as noise and high dimensional data. The relations by Bayesian network of WL are more concise and stronger in affinity too. CONCLUSION: The experiments results deliver positive signals that WL is possible to output both high classification accuracy and explainable relations graph between features and class.

Novel venom-based peptides (P13 and its derivative-M6) to maintain self-renewal of human embryonic stem cells by activating FGF and TGFß signaling pathways.

BACKGROUND: In our previous study, a venom-based peptide named Gonearrestide (also named P13) was identified and demonstrated with an effective inhibition in the proliferation of colon cancer cells. In this study, we explored if P13 and its potent mutant M6 could promote the proliferation of human embryonic stem cells and even maintain their self-renewal. METHODS: The structure-function relationship analysis on P13 and its potent mutant M6 were explored from the molecular mechanism of corresponding receptor activation by a series of inhibitor assay plus molecular and dynamics simulation studies. RESULTS: An interesting phenomenon is that P13 (and its potent mutant M6), an 18AA short peptide, can activate both FGF and TGFß signaling pathways. We demonstrated that the underlying molecular mechanisms of P13 and M6 could cooperate with proteoglycans to complete the "dimerization" of FGFR and TGFß receptors. CONCLUSIONS: Taken together, this study is the first research finding on a venom-based peptide that works on the FGF and TGF-ß signaling pathways to maintain the self-renewal of hESCs.

Deep-AmPEP30: Improve Short Antimicrobial Peptides Prediction with Deep Learning.

Antimicrobial peptides (AMPs) are a valuable source of antimicrobial agents and a potential solution to the multi-drug resistance problem. In particular, short-length AMPs have been shown to have enhanced antimicrobial activities, higher stability, and lower toxicity to human cells. We present a short-length (≤30 aa) AMP prediction method, Deep-AmPEP30, developed based on an optimal feature set of PseKRAAC reduced amino acids composition and convolutional neural network. On a balanced benchmark dataset of 188 samples, Deep-AmPEP30 yields an improved performance of 77% in accuracy, 85% in the area under the receiver operating characteristic curve (AUC-ROC), and 85% in area under the precision-recall curve (AUC-PR) over existing machine learning-based methods. To demonstrate its power, we screened the genome sequence of Candida glabrata-a gut commensal fungus expected to interact with and/or inhibit other microbes in the gut-for potential AMPs and identified a peptide of 20 aa (P3, FWELWKFLKSLWSIFPRRRP) with strong anti-bacteria activity against Bacillus subtilis and Vibrio parahaemolyticus. The potency of the peptide is remarkably comparable to that of ampicillin. Therefore, Deep-AmPEP30 is a promising prediction tool to identify short-length AMPs from genomic sequences for drug discovery. Our method is available at https://cbbio.cis.um.edu.mo/AxPEP for both individual sequence prediction and genome screening for AMPs.

Machine Learning Approaches for Quality Assessment of Protein Structures.

Protein structures play a very important role in biomedical research, especially in drug discovery and design, which require accurate protein structures in advance. However, experimental determinations of protein structure are prohibitively costly and time-consuming, and computational predictions of protein structures have not been perfected. Methods that assess the quality of protein models can help in selecting the most accurate candidates for further work. Driven by this demand, many structural bioinformatics laboratories have developed methods for estimating model accuracy (EMA). In recent years, EMA by machine learning (ML) have consistently ranked among the top-performing methods in the community-wide CASP challenge. Accordingly, we systematically review all the major ML-based EMA methods developed within the past ten years. The methods are grouped by their employed ML approach-support vector machine, artificial neural networks, ensemble learning, or Bayesian learning-and their significances are discussed from a methodology viewpoint. To orient the reader, we also briefly describe the background of EMA, including the CASP challenge and its evaluation metrics, and introduce the major ML/DL techniques. Overall, this review provides an introductory guide to modern research on protein quality assessment and directions for future research in this area.

In Vitro and MD Simulation Study to Explore Physicochemical Parameters for Antibacterial Peptide to Become Potent Anticancer Peptide.

Although the physicochemical properties of antimicrobial peptides (AMPs) and anticancer peptides (ACPs) are very similar, it remains unclear which specific parameter(s) of ACPs confer the major anticancer activity. By answering how to construct a short AMP/ACP that could easily be synthesized in the most cost effective way plus conferring a maximum anticancer effect is a very important scientific breakthrough in the development of protein/peptide drugs. In this study, an 18-amino-acids antimicrobial peptide, AcrAP1 (named AP1-Z1), was used as a template. Bioinformatics algorithms were then performed to design its six mutants (AP1-Z3a, AP1-Z3b, AP1-Z5a, AP1-Z5b, AP1-Z7, and AP1-Z9). After a series of in vitro experiments plus intensive computational analysis, the data demonstrated that AP1-Z5a and AP1-Z5b induced both apoptosis and anti-angiogenic effects to achieve the maximum anticancer activity. Specifically, the most effective mutant, AP1-Z5b, exhibited high selectivity for the charged membrane in molecular dynamics simulations. These findings clearly demonstrated that both charge and hydrophobicity play an important role and are necessary to reach an optimum equilibrium for optimizing the anticancer activity of AMPs. Overall, the present study provides a very crucial theoretical basis and important scientific evidence on the key physicochemical parameters of ACP drugs development.

Refined Empirical Force Field to Model Protein-Self-Assembled Monolayer Interactions Based on AMBER14 and GAFF.

Understanding protein interaction with material surfaces is important for the development of nanotechnological devices. The structures and dynamics of proteins can be studied via molecular dynamics (MD) if the protein-surface interactions can be accurately modeled. To answer this question, we computed the adsorption free energies of peptides (representing eleven different amino acids) on a hydrophobic self-assembled monolayer (CH3-SAM) and compared them to the benchmark experimental data set. Our result revealed that existing biomolecular force fields, GAFF and AMBER ff14sb, cannot reproduce the experimental peptide adsorption free energies by Wei and Latour (Langmuir, 2009, 25, 5637-5646). To obtain the improved force fields, we systematically tuned the Lennard-Jones parameters of selected amino acid sidechains and the functional group of SAM with repeated metadynamics and umbrella sampling simulations. The final parameter set has yielded a significant improvement in the free energy values with R = 0.83 and MSE = 0.65 kcal/mol. We applied the refined force field to predict the initial adsorption orientation of lysozyme on CH3-SAM. Two major orientations-face-down and face-up-were predicted. Our analysis on the protein structure, solvent accessible surface area, and binding of native ligand NAG3 suggested that lysozyme in the face-up orientation can remain active after initial adsorption. However, because of its weaker affinity (ΔΔG = 7.86 kcal/mol) for the ligand, the bioactivity of the protein is expected to reduce. Our work facilitates the use of MD for the study of protein-SAM systems. The refined force field compatible with GROMACS is available at https://cbbio.cis.um.edu.mo/software/SAMFF .

Structure-Based in Silico Screening Identifies a Potent Ebolavirus Inhibitor from a Traditional Chinese Medicine Library.

Potent Ebolavirus (EBOV) inhibitors will help to curtail outbreaks such as that which occurred in 2014-16 in West Africa. EBOV has on its surface a single glycoprotein (GP) critical for viral entry and membrane fusion. Recent high-resolution complexes of EBOV GP with a variety of approved drugs revealed that binding to a common cavity prevented fusion of the virus and endosomal membranes, inhibiting virus infection. We performed docking experiments, screening a database of natural compounds to identify those likely to bind at this site. Using both inhibition assays of HIV-1-derived pseudovirus cell entry and structural analyses of the complexes of the compounds with GP, we show here that two of these compounds attach in the common binding cavity, out of eight tested. In both cases, two molecules bind in the cavity. The two compounds are chemically similar, but the tighter binder has an additional chlorine atom that forms good halogen bonds to the protein and achieves an IC50 of 50 nM, making it the most potent GP-binding EBOV inhibitor yet identified, validating our screening approach for the discovery of novel antiviral compounds.

Chaos-embedded particle swarm optimization approach for protein-ligand docking and virtual screening.

BACKGROUND: Protein-ligand docking programs are routinely used in structure-based drug design to find the optimal binding pose of a ligand in the protein's active site. These programs are also used to identify potential drug candidates by ranking large sets of compounds. As more accurate and efficient docking programs are always desirable, constant efforts focus on developing better docking algorithms or improving the scoring function. Recently, chaotic maps have emerged as a promising approach to improve the search behavior of optimization algorithms in terms of search diversity and convergence speed. However, their effectiveness on docking applications has not been explored. Herein, we integrated five popular chaotic maps-logistic, Singer, sinusoidal, tent, and Zaslavskii maps-into PSOVina[Formula: see text], a recent variant of the popular AutoDock Vina program with enhanced global and local search capabilities, and evaluated their performances in ligand pose prediction and virtual screening using four docking benchmark datasets and two virtual screening datasets. RESULTS: Pose prediction experiments indicate that chaos-embedded algorithms outperform AutoDock Vina and PSOVina in ligand pose RMSD, success rate, and run time. In virtual screening experiments, Singer map-embedded PSOVina[Formula: see text] achieved a very significant five- to sixfold speedup with comparable screening performances to AutoDock Vina in terms of area under the receiver operating characteristic curve and enrichment factor. Therefore, our results suggest that chaos-embedded PSOVina methods might be a better option than AutoDock Vina for docking and virtual screening tasks. The success of chaotic maps in protein-ligand docking reveals their potential for improving optimization algorithms in other search problems, such as protein structure prediction and folding. The Singer map-embedded PSOVina[Formula: see text] which is named PSOVina-2.0 and all testing datasets are publicly available on https://cbbio.cis.umac.mo/software/psovina .