A voting-based machine learning approach for classifying biological and clinical datasets.

BACKGROUND: Different machine learning techniques have been proposed to classify a wide range of biological/clinical data. Given the practicability of these approaches accordingly, various software packages have been also designed and developed. However, the existing methods suffer from several limitations such as overfitting on a specific dataset, ignoring the feature selection concept in the preprocessing step, and losing their performance on large-size datasets. To tackle the mentioned restrictions, in this study, we introduced a machine learning framework consisting of two main steps. First, our previously suggested optimization algorithm (Trader) was extended to select a near-optimal subset of features/genes. Second, a voting-based framework was proposed to classify the biological/clinical data with high accuracy. To evaluate the efficiency of the proposed method, it was applied to 13 biological/clinical datasets, and the outcomes were comprehensively compared with the prior methods. RESULTS: The results demonstrated that the Trader algorithm could select a near-optimal subset of features with a significant level of p-value < 0.01 relative to the compared algorithms. Additionally, on the large-sie datasets, the proposed machine learning framework improved prior studies by ~ 10% in terms of the mean values associated with fivefold cross-validation of accuracy, precision, recall, specificity, and F-measure. CONCLUSION: Based on the obtained results, it can be concluded that a proper configuration of efficient algorithms and methods can increase the prediction power of machine learning approaches and help researchers in designing practical diagnosis health care systems and offering effective treatment plans.

Structure-based drug repurposing against COVID-19 and emerging infectious diseases: methods, resources and discoveries.

To attain promising pharmacotherapies, researchers have applied drug repurposing (DR) techniques to discover the candidate medicines to combat the coronavirus disease 2019 (COVID-19) outbreak. Although many DR approaches have been introduced for treating different diseases, only structure-based DR (SBDR) methods can be employed as the first therapeutic option against the COVID-19 pandemic because they rely on the rudimentary information about the diseases such as the sequence of the severe acute respiratory syndrome coronavirus 2 genome. Hence, to try out new treatments for the disease, the first attempts have been made based on the SBDR methods which seem to be among the proper choices for discovering the potential medications against the emerging and re-emerging infectious diseases. Given the importance of SBDR approaches, in the present review, well-known SBDR methods are summarized, and their merits are investigated. Then, the databases and software applications, utilized for repurposing the drugs against COVID-19, are introduced. Besides, the identified drugs are categorized based on their targets. Finally, a comparison is made between the SBDR approaches and other DR methods, and some possible future directions are proposed.

mRNA and microRNA selection for breast cancer molecular subtype stratification using meta-heuristic based algorithms.

Cancer subtype stratification, which may help to make a better decision in treating cancerous patients, is one of the most crucial and challenging problems in cancer studies. To this end, various computational methods such as Feature selection, which enhances the accuracy of the classification and is an NP-Hard problem, have been proposed. However, the performance of the applied methods is still low and can be increased by the state-of-the-art and efficient methods. We used 11 efficient and popular meta-heuristic algorithms including WCC, LCA, GA, PSO, ACO, ICA, LA, HTS, FOA, DSOS and CUK along with SVM classifier to stratify human breast cancer molecular subtypes using mRNA and micro-RNA expression data. The applied algorithms select 186 mRNAs and 116 miRNAs out of 9692 mRNAs and 489 miRNAs, respectively. Although some of the selected mRNAs and miRNAs are common in different algorithms results, six miRNAs including miR-190b, miR-18a, miR-301a, miR-34c-5p, miR-18b, and miR-129-5p were selected by equal or more than three different algorithms. Further, six mRNAs, including HAUS6, LAMA2, TSPAN33, PLEKHM3, GFRA3, and DCBLD2, were chosen through two different algorithms. We have reported these miRNAs and mRNAs as important diagnostic biomarkers to the stratification of breast cancer subtypes. By investigating the literature, it is also observed that most of our reported mRNAs and miRNAs have been proposed and introduced as biomarkers in cancer subtypes stratification.

Drug databases and their contributions to drug repurposing.

Drug repurposing is an interesting field in the drug discovery scope because of reducing time and cost. It is also considered as an appropriate method for finding medications for orphan and rare diseases. Hence, many researchers have proposed novel methods based on databases which contain different information. Thus, a suitable organization of data which facilitates the repurposing applications and provides a tool or a web service can be beneficial. In this review, we categorize drug databases and discuss their advantages and disadvantages. Surprisingly, to the best of our knowledge, the importance and potential of databases in drug repurposing are yet to be emphasized. Indeed, the available databases can be divided into several groups based on data content, and different classes can be applied to find a new application of the existing drugs. Furthermore, we propose some suggestions for making databases more effective and popular in this field.

Synthetic repurposing of drugs against hypertension: a datamining method based on association rules and a novel discrete algorithm.

BACKGROUND: Drug repurposing aims to detect the new therapeutic benefits of the existing drugs and reduce the spent time and cost of the drug development projects. The synthetic repurposing of drugs may prove to be more useful than the single repurposing in terms of reducing toxicity and enhancing efficacy. However, the researchers have not given it serious consideration. To address the issue, a novel datamining method is introduced and applied to repositioning of drugs for hypertension (HT) which is a serious medical condition and needs some improved treatment plans to help treat it. RESULTS: A novel two-step data mining method, which is based on the If-Then association rules as well as a novel discrete optimization algorithm, was introduced and applied to the synthetic repurposing of drugs for HT. The required data were also extracted from DrugBank, KEGG, and DrugR+ databases. The findings indicated that based on the different statistical criteria, the proposed method outperformed the other state-of-the-art approaches. In contrast to the previously proposed methods which had failed to discover a list on some datasets, our method could find a combination list for all of them. CONCLUSION: Since the proposed synthetic method uses medications in small dosages, it might revive some failed drug development projects and put forward a suitable plan for treating different diseases such as COVID-19 and HT. It is also worth noting that applying efficient computational methods helps to produce better results.

FeatureSelect: a software for feature selection based on machine learning approaches.

BACKGROUND: Feature selection, as a preprocessing stage, is a challenging problem in various sciences such as biology, engineering, computer science, and other fields. For this purpose, some studies have introduced tools and softwares such as WEKA. Meanwhile, these tools or softwares are based on filter methods which have lower performance relative to wrapper methods. In this paper, we address this limitation and introduce a software application called FeatureSelect. In addition to filter methods, FeatureSelect consists of optimisation algorithms and three types of learners. It provides a user-friendly and straightforward method of feature selection for use in any kind of research, and can easily be applied to any type of balanced and unbalanced data based on several score functions like accuracy, sensitivity, specificity, etc. RESULTS: In addition to our previously introduced optimisation algorithm (WCC), a total of 10 efficient, well-known and recently developed algorithms have been implemented in FeatureSelect. We applied our software to a range of different datasets and evaluated the performance of its algorithms. Acquired results show that the performances of algorithms are varying on different datasets, but WCC, LCA, FOA, and LA are suitable than others in the overall state. The results also show that wrapper methods are better than filter methods. CONCLUSIONS: FeatureSelect is a feature or gene selection software application which is based on wrapper methods. Furthermore, it includes some popular filter methods and generates various comparison diagrams and statistical measurements. It is available from GitHub ( https://github.com/LBBSoft/FeatureSelect ) and is free open source software under an MIT license.

DNA polymerase ε and δ variants drive mutagenesis in polypurine tracts in human tumors.

Alterations in the exonuclease domain of DNA polymerase ε cause ultramutated cancers. These cancers accumulate AGA>ATA transversions; however, their genomic features beyond the trinucleotide motifs are obscure. We analyze the extended DNA context of ultramutation using whole-exome sequencing data from 524 endometrial and 395 colorectal tumors. We find that G>T transversions in POLE-mutant tumors predominantly affect sequences containing at least six consecutive purines, with a striking preference for certain positions within polypurine tracts. Using this signature, we develop a machine-learning classifier to identify tumors with hitherto unknown POLE drivers and validate two drivers, POLE-E978G and POLE-S461L, by functional assays in yeast. Unlike other pathogenic variants, the E978G substitution affects the polymerase domain of Pol ε. We further show that tumors with POLD1 drivers share the extended signature of POLE ultramutation. These findings expand the understanding of ultramutation mechanisms and highlight peculiar mutagenic properties of polypurine tracts in the human genome.

Tracing drugs from discovery to disposal.

The life cycle of a drug begins with discovery and ends with its disposal. Drug discovery companies, drug manufacturers, regulatory agencies, suppliers, pharmacies, patients, healthcare providers, and many more are involved in this process. Transparency, traceability, automation, and data security are some of the most crucial factors affecting how effectively and safely the transactions are conducted across all parties involved in the cycle. By contrast, scalability, energy consumption, regulation, standards, and complexity hamper the adoption of new technology that is expected to fulfil these requirements. Here, we highlight how blockchain technology can track, accelerate, and boost the efficiency of incredibly complicated operations, such as pharmaceutical development.

Deciphering anti-biofilm property of Arthrospira platensis-origin peptides against Staphylococcusaureus.

Arthrospira platensis is a valuable natural health supplement consisting of various types of vitamins, dietary minerals, and antioxidants. Although different studies have been conducted to explore the hidden benefits of this bacterium, its antimicrobial property has been poorly understood. To decipher this important feature, here, we extended our recently introduced optimization algorithm (Trader) for aligning amino acid sequences associated with the antimicrobial peptides (AMPs) of Staphylococcus aureus and A.platensis. As a result, similar amino acid sequences were identified, and several candidate peptides were generated accordingly. The obtained peptides were then filtered based on their potential biochemical and biophysical properties, and their 3D structures were simulated based on homology modeling techniques. Next, to investigate how the generated peptides can interact with S. aureus proteins (i.e., heptameric state of the hly and homodimeric form of the arsB), molecular docking approaches were used. The results indicated that four peptides included better molecular interactions relative to the other generated ones in terms of the number/average length of hydrogen bonds and hydrophobic interactions. Based on the outcomes, it can be concluded that the antimicrobial property of A.platensis might be associated with its capability in disturbing the membrane of pathogens and their functions.

A fuzzy logic-based computational method for the repurposing of drugs against COVID-19.

Introduction: COVID-19 has spread out all around the world and seriously interrupted human activities. Being a newfound disease, not only many aspects of the disease are unknown, but also there is not an effective medication to cure the disease. Besides, designing a drug is a time-consuming process and needs large investment. Hence, drug repurposing techniques, employed to discover the hidden benefits of the existing drugs, maybe a useful option for treating COVID-19. Methods: The present study exploits the drug repositioning concepts and introduces some candidate drugs which may be effective in controlling COVID-19. The suggested method consists of three main steps. First, the required data such as the amino acid sequences of targets and drug-target interactions are extracted from the public databases. Second, the similarity score between the targets (protein/enzymes) and genome of SARS-COV-2 is computed using the proposed fuzzy logic-based method. Since the classical approaches yield outcomes which may not be useful for the real-world applications, the fuzzy technique can address the issue. Third, after ranking targets based on the obtained scores, the usefulness of drugs affecting them is examined for managing COVID-19. Results: The results indicate that antiviral medicines, designed for curing hepatitis C, may also cure COVID-19. According to the findings, ribavirin, simeprevir, danoprevir, and XTL-6865 may be helpful in controlling the disease. Conclusion: It can be concluded that the similarity-based drug repurposing techniques may be the most suitable option for managing emerging diseases such as COVID-19 and can be applied to a wide range of data. Also, fuzzy logic-based scoring methods can produce outcomes which are more consistent with the real-world biological applications than others.

Discovering driver nodes in chronic kidney disease-related networks using Trader as a newly developed algorithm.

Thanks to the advances in the field of computational-based biology, a huge volume of disease-related data has been generated so far. From the existing data, the disease-related protein-protein interaction (PPI) networks seem to yield effective treatment plans due to the informative/systematic representation of diseases. Yet, a large number of previous studies have failed due to the complex nature of such disease-related networks. For addressing this limitation, in the present study, we combined Trader and the DFS algorithms to identify a minimal subset of nodes (driver nodes) whose removal produces a maximum number of disjoint sub-networks. We then screened the nodes in the disease-associated PPI networks and to evaluate the efficiency of the suggested method, it was applied to six PPI networks of differentially expressed genes in chronic kidney diseases. The performance of Trader was superior to other well-known algorithms in terms of identifying driver nodes. Besides, the proportion of proteins that were targeted by at least one FDA-approved drug was significantly higher among the identified driver nodes when compared with the rest of the proteins in the networks. The proposed algorithm could be applied for predicting future therapeutic targets in complex disorder networks. In conclusion, unlike the common methods, computationally efficient algorithms can generate more practical outcomes which are compatible with real-world biological facts.

Self-amplifying mRNA vaccines: Mode of action, design, development and optimization.

The mRNA-based vaccines are quality-by-design (QbD) immunotherapies that provide safe, tunable, scalable, streamlined and potent treatment possibilities against different types of diseases. The self-amplifying mRNA (saRNA) vaccines, as a highly advantageous class of mRNA vaccines, are inspired by the intracellular self-multiplication nature of some positive-sense RNA viruses. Such vaccine platforms provide a relatively increased expression level of vaccine antigen(s) together with self-adjuvanticity properties. Lined with the QbD saRNA vaccines, essential optimizations improve the stability, safety, and immunogenicity of the vaccine constructs. Here, we elaborate on the concepts and mode-of-action of mRNA and saRNA vaccines, articulate the potential limitations or technical bottlenecks, and explain possible solutions or optimization methods in the process of their design and development.

Advances in the computational landscape for repurposed drugs against COVID-19.

The COVID-19 pandemic has caused millions of deaths and massive societal distress worldwide. Therapeutic solutions are urgently needed, but de novo drug development remains a lengthy process. One promising alternative is computational drug repurposing, which enables the prioritization of existing compounds through fast in silico analyses. Recent efforts based on molecular docking, machine learning, and network analysis have produced actionable predictions. Some predicted drugs, targeting viral proteins and pathological host pathways are undergoing clinical trials. Here, we review this work, highlight drugs with high predicted efficacy and classify their mechanisms of action. We discuss the strengths and limitations of the published methodologies and outline possible future directions. Finally, we curate a list of COVID-19 data portals and other repositories that could be used to accelerate future research.

A machine learning method based on the genetic and world competitive contests algorithms for selecting genes or features in biological applications.

Gene/feature selection is an essential preprocessing step for creating models using machine learning techniques. It also plays a critical role in different biological applications such as the identification of biomarkers. Although many feature/gene selection algorithms and methods have been introduced, they may suffer from problems such as parameter tuning or low level of performance. To tackle such limitations, in this study, a universal wrapper approach is introduced based on our introduced optimization algorithm and the genetic algorithm (GA). In the proposed approach, candidate solutions have variable lengths, and a support vector machine scores them. To show the usefulness of the method, thirteen classification and regression-based datasets with different properties were chosen from various biological scopes, including drug discovery, cancer diagnostics, clinical applications, etc. Our findings confirmed that the proposed method outperforms most of the other currently used approaches and can also free the users from difficulties related to the tuning of various parameters. As a result, users may optimize their biological applications such as obtaining a biomarker diagnostic kit with the minimum number of genes and maximum separability power.

Pharmacoinformatics-based phytochemical screening for anticancer impacts of yellow sweet clover, Melilotus officinalis (Linn.) Pall.

To date, much attention has been paid to phytochemicals because of their diverse pharmacological effects on a variety of diseases such as cancer. In this regard, computer-aided drug design, as a cost- and time-effective approach, is primarily applied to investigate the drug candidates before their further costly in vitro and in vivo experimental evaluations. Accordingly, different signaling pathways and proteins can be targeted using such strategies. As a key protein for the initiation of eukaryotic DNA replication, mini-chromosome maintenance complex component 7 (MCM7) overexpression is related to the initiation and progression of aggressive malignancies. The current study was conducted to identify new potential natural compounds from the yellow sweet clover, Melilotus officinalis (Linn.) Pall, by examining the potential of 40 isolated phytochemicals against MCM7 protein. A structure-based pharmacophore model to the protein active site cavity was generated and followed by virtual screening and molecular docking. Overall, four compounds were selected for further evaluation based on their binding affinities. Our analyses revealed that two novel compounds, namely rosmarinic acid (PubChem CID:5281792) and melilotigenin (PubChem CID:14059499) might be druggable and offer safe usage in human. The stability of these two protein-ligand complex structures was confirmed through molecular dynamics simulation. The findings of this study reveal the potential of these two phytochemicals to serve as anticancer agents, while further pharmacological experiments are required to confirm their effectiveness against human cancers.

A novel multi-objective metaheuristic algorithm for protein-peptide docking and benchmarking on the LEADS-PEP dataset.

Protein-peptide interactions have attracted the attention of many drug discovery scientists due to their possible druggability features on most key biological activities such as regulating disease-related signaling pathways and enhancing the immune system's responses. Different studies have utilized some protein-peptide-specific docking algorithms/methods to predict protein-peptide interactions. However, the existing algorithms/methods suffer from two serious limitations which make them unsuitable for protein-peptide docking problems. First, it seems that the prevalent approaches require to be modified and remodeled for weighting the unbounded forces between a protein and a peptide. Second, they do not employ state-of-the-art search algorithms for detecting the 3D pose of a peptide relative to a protein. To address these restrictions, the present study aims to introduce a novel multi-objective algorithm, which first generates some potential 3D poses of a peptide, and then, improves them through its operators. The candidate solutions are further evaluated using Multi-Objective Pareto Front (MOPF) optimization concepts. To this end, van der Waals, electrostatic, solvation, and hydrogen bond energies between the atoms of a protein and designated peptide are computed. To evaluate the algorithm, it is first applied to the LEADS-PEP dataset containing 53 protein-peptide complexes with up to 53 rotatable branches/bonds and then compared with three popular/efficient algorithms. The obtained results indicate that the MOPF-based approaches which reduce the backbone RMSD between the original and predicted states, achieve significantly better results in terms of the success rate in predicting the near-native conditions. Besides, a comparison between the different types of search algorithms reveals that efficient ones like the multi-objective Trader/differential evolution algorithm can predict protein-peptide interactions better than the popular algorithms such as the multi-objective genetic/particle swarm optimization algorithms.

Computational-based drug repurposing methods in COVID-19.

COVID-19, as a newly emerging disease, has disrupted human's different activities. Hence, it is essential to develop drugs or vaccines in order to control COVID-19. Since there is not a medication or vaccine for treating the disease and drug development project is a time and cost consuming process, drug repurposing approaches may yield to proper curing plans. However, there are some limitations in this field, which make the process a challenging one. This letter aims to introduce drug repurposing methods and the existing challenges to detect candidate drugs which may be helpful in controlling COVID-19.

Trader as a new optimization algorithm predicts drug-target interactions efficiently.

Several machine learning approaches have been proposed for predicting new benefits of the existing drugs. Although these methods have introduced new usage(s) of some medications, efficient methods can lead to more accurate predictions. To this end, we proposed a novel machine learning method which is based on a new optimization algorithm, named Trader. To show the capabilities of the proposed algorithm which can be applied to the different scope of science, it was compared with ten other state-of-the-art optimization algorithms based on the standard and advanced benchmark functions. Next, a multi-layer artificial neural network was designed and trained by Trader to predict drug-target interactions (DTIs). Finally, the functionality of the proposed method was investigated on some DTIs datasets and compared with other methods. The data obtained by Trader showed that it eliminates the disadvantages of different optimization algorithms, resulting in a better outcome. Further, the proposed machine learning method was found to achieve a significant level of performance compared to the other popular and efficient approaches in predicting unknown DTIs. All the implemented source codes are freely available at https://github.com/LBBSoft/Trader .

DrugR+: A comprehensive relational database for drug repurposing, combination therapy, and replacement therapy.

Drug repurposing or repositioning, which introduces new applications of the existing drugs, is an emerging field in drug discovery scope. To enhance the success rate of the research and development (R&D) process in a cost- and time-effective manner, a number of pharmaceutical companies worldwide have made tremendous investments. Besides, many researchers have proposed various methods and databases for the repurposing of various drugs. However, there is not a proper and well-organized database available. To this end, for the first time, we developed a new database based on DrugBank and KEGG data, which is named "DrugR+". Our developed database provides some advantages relative to the DrugBank, and its interface supplies new capabilities for both single and synthetic repositioning of drugs. Moreover, it includes four new datasets which can be used for predicting drug-target interactions using supervised machine learning methods. As a case study, we introduced novel applications of some drugs and discussed the obtained results. A comparison of several machine learning methods on the generated datasets has also been reported in the Supplementary File. Having included several normalized tables, DrugR + has been organized to provide key information on data structures for the repurposing and combining applications of drugs. It provides the SQL query capability for professional users and an appropriate method with different options for unprofessional users. Additionally, DrugR + consists of repurposing service that accepts a drug and proposes a list of potential drugs for some usages. Taken all, DrugR+ is a free web-based database and accessible using (http://www.drugr.ir), which can be updated through a map-reduce parallel processing method to provide the most relevant information.

Sequential and Mixed Genetic Algorithm and Learning Automata (SGALA, MGALA) for Feature Selection in QSAR.

Feature selection is of great importance in Quantitative Structure-Activity Relationship (QSAR) analysis. This problem has been solved using some meta-heuristic algorithms such as GA, PSO, ACO and so on. In this work two novel hybrid meta-heuristic algorithms i.e. Sequential GA and LA (SGALA) and Mixed GA and LA (MGALA), which are based on Genetic algorithm and learning automata for QSAR feature selection are proposed. SGALA algorithm uses advantages of Genetic algorithm and Learning Automata sequentially and the MGALA algorithm uses advantages of Genetic Algorithm and Learning Automata simultaneously. We applied our proposed algorithms to select the minimum possible number of features from three different datasets and also we observed that the MGALA and SGALA algorithms had the best outcome independently and in average compared to other feature selection algorithms. Through comparison of our proposed algorithms, we deduced that the rate of convergence to optimal result in MGALA and SGALA algorithms were better than the rate of GA, ACO, PSO and LA algorithms. In the end, the results of GA, ACO, PSO, LA, SGALA, and MGALA algorithms were applied as the input of LS-SVR model and the results from LS-SVR models showed that the LS-SVR model had more predictive ability with the input from SGALA and MGALA algorithms than the input from all other mentioned algorithms. Therefore, the results have corroborated that not only is the predictive efficiency of proposed algorithms better, but their rate of convergence is also superior to the all other mentioned algorithms.