MEvA-X: a hybrid multiobjective evolutionary tool using an XGBoost classifier for biomarkers discovery on biomedical datasets.

MOTIVATION: Biomarker discovery is one of the most frequent pursuits in bioinformatics and is crucial for precision medicine, disease prognosis, and drug discovery. A common challenge of biomarker discovery applications is the low ratio of samples over features for the selection of a reliable not-redundant subset of features, but despite the development of efficient tree-based classification methods, such as the extreme gradient boosting (XGBoost), this limitation is still relevant. Moreover, existing approaches for optimizing XGBoost do not deal effectively with the class imbalance nature of the biomarker discovery problems, and the presence of multiple conflicting objectives, since they focus on the training of a single-objective model. In the current work, we introduce MEvA-X, a novel hybrid ensemble for feature selection (FS) and classification, combining a niche-based multiobjective evolutionary algorithm (EA) with the XGBoost classifier. MEvA-X deploys a multiobjective EA to optimize the hyperparameters of the classifier and perform FS, identifying a set of Pareto-optimal solutions and optimizing multiple objectives, including classification and model simplicity metrics. RESULTS: The performance of the MEvA-X tool was benchmarked using one omics dataset coming from a microarray gene expression experiment, and one clinical questionnaire-based dataset combined with demographic information. MEvA-X tool outperformed the state-of-the-art methods in the balanced categorization of classes, creating multiple low-complexity models and identifying important nonredundant biomarkers. The best-performing run of MEvA-X for the prediction of weight loss using gene expression data yields a small set of blood circulatory markers which are sufficient for this precision nutrition application but need further validation. AVAILABILITY AND IMPLEMENTATION: https://github.com/PanKonstantinos/MEvA-X.

Machine learning approaches to predict TAS2R receptors for bitterants.

Bitter taste involves the detection of diverse chemical compounds by a family of G protein-coupled receptors, known as taste receptor type 2 (TAS2R). It is often linked to toxins and harmful compounds and in particular bitter taste receptors participate in the regulation of glucose homeostasis, modulation of immune and inflammatory responses, and may have implications for various diseases. Human TAS2Rs are characterized by their polymorphism and differ in localization and function. Different receptors can activate various signaling pathways depending on the tissue and the ligand. However, in vitro screening of possible TAS2R ligands is costly and time-consuming. For this reason, in silico methods to predict bitterant-TAS2R interactions could be powerful tools to help in the selection of ligands and targets for experimental studies and improve our knowledge of bitter receptor roles. Machine learning (ML) is a branch of artificial intelligence that applies algorithms to large datasets to learn from patterns and make predictions. In recent years, there has been a record of numerous taste classifiers in literature, especially on bitter/non-bitter or bitter/sweet classification. However, only a few of them exploit ML to predict which TAS2R receptors could be targeted by bitter molecules. Indeed, the shortage and incompleteness of data on receptor-ligand associations in literature make this task non-trivial. In this work, we provide an overview of the state of the art dealing with this specific investigation, focusing on three ML-based models, namely BitterX (2016), BitterSweet (2019) and BitterMatch (2022). This review aims to establish the foundation for future research endeavours focused on addressing the limitations and drawbacks of existing models.

Validation of Step Detection and Distance Calculation Algorithms for Soccer Performance Monitoring.

This study focused on developing and evaluating a gyroscope-based step counter algorithm using inertial measurement unit (IMU) readings for precise athletic performance monitoring in soccer. The research aimed to provide reliable step detection and distance estimation tailored to soccer-specific movements, including various running speeds and directional changes. Real-time algorithms utilizing shank angular data from gyroscopes were created. Experiments were conducted on a specially designed soccer-specific testing circuit performed by 15 athletes, simulating a range of locomotion activities such as walking, jogging, and high-intensity actions. The algorithm outcome was compared with manually tagged data from a high-quality video camera-based system for validation, by assessing the agreement between the paired values using limits of agreement, concordance correlation coefficient, and further metrics. Results returned a step detection accuracy of 95.8% and a distance estimation Root Mean Square Error (RMSE) of 17.6 m over about 202 m of track. A sub-sample (N = 6) also wore two pairs of devices concurrently to evaluate inter-unit reliability. The performance analysis suggested that the algorithm was effective and reliable in tracking diverse soccer-specific movements. The proposed algorithm offered a robust and efficient solution for tracking step count and distance covered in soccer, particularly beneficial in indoor environments where global navigation satellite systems are not feasible. This advancement in sports technology widens the spectrum of tools for coaches and athletes in monitoring soccer performance.

Toward the design and development of peptidomimetic inhibitors of the Ataxin-1 aggregation pathway.

Spinocerebellar ataxia type 1 is a degenerative disorder caused by polyglutamine expansions and aggregation of Ataxin-1. The interaction between Capicua (CIC) and the AXH domain of Ataxin-1 protein has been suggested as a possible driver of aggregation for the expanded Ataxin-1 protein and the subsequent onset of spinocerebellar ataxia 1. Experimental studies have demonstrated that short constructs of CIC may prevent such aggregation and suggested this as a possible candidate to inspire the rational design of peptidomimetics. In this work, molecular modeling techniques, namely the alchemical mutation and force field-based molecular dynamics, have been employed to propose a pipeline for the rational design of a CIC-inspired inhibitor of the ataxin-1 aggregation pathway. In particular, this study has shown that the alchemical mutation can estimate the affinity between AXH and CIC with good correlation with experimental data, while molecular dynamics shed light on molecular mechanisms that occur for stabilization of the interaction between the CIC-inspired construct and the AXH domain of Ataxin-1. This work lays the foundation for a rational methodology for the in silico screening and design of peptidomimetics, which can expedite and streamline experimental studies to identify strategies for inhibiting the ataxin-1 aggregation pathway.

Machine learning-based prediction of adverse events following an acute coronary syndrome (PRAISE): a modelling study of pooled datasets.

BACKGROUND: The accuracy of current prediction tools for ischaemic and bleeding events after an acute coronary syndrome (ACS) remains insufficient for individualised patient management strategies. We developed a machine learning-based risk stratification model to predict all-cause death, recurrent acute myocardial infarction, and major bleeding after ACS. METHODS: Different machine learning models for the prediction of 1-year post-discharge all-cause death, myocardial infarction, and major bleeding (defined as Bleeding Academic Research Consortium type 3 or 5) were trained on a cohort of 19 826 adult patients with ACS (split into a training cohort [80%] and internal validation cohort [20%]) from the BleeMACS and RENAMI registries, which included patients across several continents. 25 clinical features routinely assessed at discharge were used to inform the models. The best-performing model for each study outcome (the PRAISE score) was tested in an external validation cohort of 3444 patients with ACS pooled from a randomised controlled trial and three prospective registries. Model performance was assessed according to a range of learning metrics including area under the receiver operating characteristic curve (AUC). FINDINGS: The PRAISE score showed an AUC of 0·82 (95% CI 0·78-0·85) in the internal validation cohort and 0·92 (0·90-0·93) in the external validation cohort for 1-year all-cause death; an AUC of 0·74 (0·70-0·78) in the internal validation cohort and 0·81 (0·76-0·85) in the external validation cohort for 1-year myocardial infarction; and an AUC of 0·70 (0·66-0·75) in the internal validation cohort and 0·86 (0·82-0·89) in the external validation cohort for 1-year major bleeding. INTERPRETATION: A machine learning-based approach for the identification of predictors of events after an ACS is feasible and effective. The PRAISE score showed accurate discriminative capabilities for the prediction of all-cause death, myocardial infarction, and major bleeding, and might be useful to guide clinical decision making. FUNDING: None.

A Machine Learning Approach for Mortality Prediction in COVID-19 Pneumonia: Development and Evaluation of the Piacenza Score.

BACKGROUND: Several models have been developed to predict mortality in patients with COVID-19 pneumonia, but only a few have demonstrated enough discriminatory capacity. Machine learning algorithms represent a novel approach for the data-driven prediction of clinical outcomes with advantages over statistical modeling. OBJECTIVE: We aimed to develop a machine learning-based score-the Piacenza score-for 30-day mortality prediction in patients with COVID-19 pneumonia. METHODS: The study comprised 852 patients with COVID-19 pneumonia, admitted to the Guglielmo da Saliceto Hospital in Italy from February to November 2020. Patients' medical history, demographics, and clinical data were collected using an electronic health record. The overall patient data set was randomly split into derivation and test cohorts. The score was obtained through the naïve Bayes classifier and externally validated on 86 patients admitted to Centro Cardiologico Monzino (Italy) in February 2020. Using a forward-search algorithm, 6 features were identified: age, mean corpuscular hemoglobin concentration, PaO2/FiO2 ratio, temperature, previous stroke, and gender. The Brier index was used to evaluate the ability of the machine learning model to stratify and predict the observed outcomes. A user-friendly website was designed and developed to enable fast and easy use of the tool by physicians. Regarding the customization properties of the Piacenza score, we added a tailored version of the algorithm to the website, which enables an optimized computation of the mortality risk score for a patient when some of the variables used by the Piacenza score are not available. In this case, the naïve Bayes classifier is retrained over the same derivation cohort but using a different set of patient characteristics. We also compared the Piacenza score with the 4C score and with a naïve Bayes algorithm with 14 features chosen a priori. RESULTS: The Piacenza score exhibited an area under the receiver operating characteristic curve (AUC) of 0.78 (95% CI 0.74-0.84, Brier score=0.19) in the internal validation cohort and 0.79 (95% CI 0.68-0.89, Brier score=0.16) in the external validation cohort, showing a comparable accuracy with respect to the 4C score and to the naïve Bayes model with a priori chosen features; this achieved an AUC of 0.78 (95% CI 0.73-0.83, Brier score=0.26) and 0.80 (95% CI 0.75-0.86, Brier score=0.17), respectively. CONCLUSIONS: Our findings demonstrated that a customizable machine learning-based score with a purely data-driven selection of features is feasible and effective for the prediction of mortality among patients with COVID-19 pneumonia.

Using the Gibbs Function as a Measure of Human Brain Development Trends from Fetal Stage to Advanced Age.

We propose to use a Gibbs free energy function as a measure of the human brain development. We adopt this approach to the development of the human brain over the human lifespan: from a prenatal stage to advanced age. We used proteomic expression data with the Gibbs free energy to quantify human brain's protein-protein interaction networks. The data, obtained from BioGRID, comprised tissue samples from the 16 main brain areas, at different ages, of 57 post-mortem human brains. We found a consistent functional dependence of the Gibbs free energies on age for most of the areas and both sexes. A significant upward trend in the Gibbs function was found during the fetal stages, which is followed by a sharp drop at birth with a subsequent period of relative stability and a final upward trend toward advanced age. We interpret these data in terms of structure formation followed by its stabilization and eventual deterioration. Furthermore, gender data analysis has uncovered the existence of functional differences, showing male Gibbs function values lower than female at prenatal and neonatal ages, which become higher at ages 8 to 40 and finally converging at late adulthood with the corresponding female Gibbs functions.

The Impact of Natural Compounds on S-Shaped Aß42 Fibril: From Molecular Docking to Biophysical Characterization.

The pursuit for effective strategies inhibiting the amyloidogenic process in neurodegenerative disorders, such as Alzheimer's disease (AD), remains one of the main unsolved issues, and only a few drugs have demonstrated to delay the degeneration of the cognitive system. Moreover, most therapies induce severe side effects and are not effective at all stages of the illness. The need to find novel and reliable drugs appears therefore of primary importance. In this context, natural compounds have shown interesting beneficial effects on the onset and progression of neurodegenerative diseases, exhibiting a great inhibitory activity on the formation of amyloid aggregates and proving to be effective in many preclinical and clinical studies. However, their inhibitory mechanism is still unclear. In this work, ensemble docking and molecular dynamics simulations on S-shaped Aß42 fibrils have been carried out to evaluate the influence of several natural compounds on amyloid conformational behaviour. A deep understanding of the interaction mechanisms between natural compounds and Aß aggregates may play a key role to pave the way for design, discovery and optimization strategies toward an efficient destabilization of toxic amyloid assemblies.

Fludarabine-Specific Molecular Interactions with Maltose-Modified Poly(propyleneimine) Dendrimer Enable Effective Cell Entry of the Active Drug Form: Comparison with Clofarabine.

Fludarabine is an anticancer antimetabolite essential for modern chemotherapy, but its efficacy is limited due to the complex pharmacokinetics. We demonstrated the potential use of maltose-modified poly(propyleneimine) dendrimer as drug delivery agent to improve the efficiency of therapy with fludarabine. In this study, we elaborated a novel synthesis technique for radioactively labeled fludarabine triphosphate to prove for the first time the direct ability of nucleotide-glycodendrimer complex to enter and kill leukemic cells, without the involvement of membrane nucleoside transporters and intracellular kinases. This will potentially allow to bypass the most common drug resistance mechanisms observed in the clinical setting. Further, we applied surface plasmon resonance and molecular modeling to elucidate the properties of the drug-dendrimer complexes. We showed that clofarabine, a more toxic nucleoside analogue drug, is characterized by significantly different molecular interactions with poly(propyleneimine) dendrimers than fludarabine, leading to different cellular outcomes (decreased rather than increased treatment efficiency). The most probable mechanistic explanation of uniquely dendrimer-enhanced fludarabine toxicity points to a crucial role of both an alternative cellular uptake pathway and the avoidance of intracellular phosphorylation of nucleoside drug form.

The synergistic effect of chlorotoxin-mApoE in boosting drug-loaded liposomes across the BBB.

We designed liposomes dually functionalized with ApoE-derived peptide (mApoE) and chlorotoxin (ClTx) to improve their blood-brain barrier (BBB) crossing. Our results demonstrated the synergistic activity of ClTx-mApoE in boosting doxorubicin-loaded liposomes across the BBB, keeping the anti-tumour activity of the drug loaded: mApoE acts promoting cellular uptake, while ClTx promotes exocytosis of liposomes.

Destabilizing the AXH Tetramer by Mutations: Mechanisms and Potential Antiaggregation Strategies.

The AXH domain of protein Ataxin 1 is thought to play a key role in the misfolding and aggregation pathway responsible for Spinocerebellar ataxia 1. For this reason, a molecular level understanding of AXH oligomerization pathway is crucial to elucidate the aggregation mechanism, which is thought to trigger the disease. This study employs classical and enhanced molecular dynamics to identify the structural and energetic basis of AXH tetramer stability. Results of this work elucidate molecular mechanisms behind the destabilizing effect of protein mutations, which consequently affect the AXH tetramer assembly. Moreover, results of the study draw attention for the first time, to our knowledge, to the R638 protein residue, which is shown to play a key role in AXH tetramer stability. Therefore, R638 might be also implicated in the AXH oligomerization pathway and stands out as a target for future experimental studies focused on self-association mechanisms and fibril formation of full-length ATX1.

Protein Environment: A Crucial Triggering Factor in Josephin Domain Aggregation: The Role of 2,2,2-Trifluoroethanol.

The protein ataxin-3 contains a polyglutamine stretch that triggers amyloid aggregation when it is expanded beyond a critical threshold. This results in the onset of the spinocerebellar ataxia type 3. The protein consists of the globular N-terminal Josephin domain and a disordered C-terminal tail where the polyglutamine stretch is located. Expanded ataxin-3 aggregates via a two-stage mechanism: first, Josephin domain self-association, then polyQ fibrillation. This highlights the intrinsic amyloidogenic potential of Josephin domain. Therefore, much effort has been put into investigating its aggregation mechanism(s). A key issue regards the conformational requirements for triggering amyloid aggregation, as it is believed that, generally, misfolding should precede aggregation. Here, we have assayed the effect of 2,2,2-trifluoroethanol, a co-solvent capable of stabilizing secondary structures, especially α-helices. By combining biophysical methods and molecular dynamics, we demonstrated that both secondary and tertiary JD structures are virtually unchanged in the presence of up to 5% 2,2,2-trifluoroethanol. Despite the preservation of JD structure, 1% of 2,2,2-trifluoroethanol suffices to exacerbate the intrinsic aggregation propensity of this domain, by slightly decreasing its conformational stability. These results indicate that in the case of JD, conformational fluctuations might suffice to promote a transition towards an aggregated state without the need for extensive unfolding, and highlights the important role played by the environment on the aggregation of this globular domain.

Conformational Dynamics and Stability of U-Shaped and S-Shaped Amyloid ß Assemblies.

Alzheimer's disease is the most fatal neurodegenerative disorder characterized by the aggregation and deposition of Amyloid ß (Aß) oligomers in the brain of patients. Two principal variants of Aß exist in humans: Aß1-40 and Aß1-42. The former is the most abundant in the plaques, while the latter is the most toxic species and forms fibrils more rapidly. Interestingly, fibrils of Aß1-40 peptides can only assume U-shaped conformations while Aß1-42 can also arrange as S-shaped three-stranded chains, as recently discovered. As alterations in protein conformational arrangement correlate with cell toxicity and speed of disease progression, it is important to characterize, at molecular level, the conformational dynamics of amyloid fibrils. In this work, Replica Exchange Molecular Dynamics simulations were carried out to compare the conformational dynamics of U-shaped and S-shaped Aß17-42 small fibrils. Our computational results provide support for the stability of the recently proposed S-shaped model due to the maximized interactions involving the C-terminal residues. On the other hand, the U-shaped motif is characterized by significant distortions resulting in a more disordered assembly. Outcomes of our work suggest that the molecular architecture of the protein aggregates might play a pivotal role in formation and conformational stability of the resulting fibrils.

Self-Assembled Ligands Targeting TLR7: A Molecular Level Investigation.

Toll-like receptors (TLRs) are pattern recognition transmembrane proteins that play an important role in innate immunity. In particular, TLR7 plays a role in detecting nucleic acids derived from viruses and bacteria. The huge number of pathologies in which TLR7 is involved has led to an increasing interest in developing new compounds targeting this protein. Several conjugation strategies were proposed for TLR7 agonists to increase the potency while maintaining a low toxicity. In this work, we focus the attention on two promising classes of TLR7 compounds derived from the same pharmacophore conjugated with phospholipid and polyethylene glycol (PEG). A multidisciplinary investigation has been carried out by molecular dynamics (MD), dynamic light scattering (DLS), electron paramagnetic resonance (EPR), and cytotoxicity assessment. DLS and MD indicated how only the phospholipid conjugation provides the compound abilities to self-assemble in an orderly fashion with a maximal pharmacophore exposition to the solvent. Further EPR and cytotoxicity experiments highlighted that phospholipid compounds organize in stable aggregates and well interact with TLR7, whereas PEG conjugation was characterized by poorly stable aggregates at the cells surface. The methodological framework proposed in this study may be used to investigate, at a molecular level, the interactions generally occurring between aggregated ligands, to be used as drugs, and protein receptors.

Josephin Domain Structural Conformations Explored by Metadynamics in Essential Coordinates.

The Josephin Domain (JD), i.e. the N-terminal domain of Ataxin 3 (At3) protein, is an interesting example of competition between physiological function and aggregation risk. In fact, the fibrillogenesis of Ataxin 3, responsible for the spinocerebbellar ataxia 3, is strictly related to the JD thermodynamic stability. Whereas recent NMR studies have demonstrated that different JD conformations exist, the likelihood of JD achievable conformational states in solution is still an open issue. Marked differences in the available NMR models are located in the hairpin region, supporting the idea that JD has a flexible hairpin in dynamic equilibrium between open and closed states. In this work we have carried out an investigation on the JD conformational arrangement by means of both classical molecular dynamics (MD) and Metadynamics employing essential coordinates as collective variables. We provide a representation of the free energy landscape characterizing the transition pathway from a JD open-like structure to a closed-like conformation. Findings of our in silico study strongly point to the closed-like conformation as the most likely for a Josephin Domain in water.

Explaining the Microtubule Energy Balance: Contributions Due to Dipole Moments, Charges, van der Waals and Solvation Energy.

Microtubules are the main components of mitotic spindles, and are the pillars of the cellular cytoskeleton. They perform most of their cellular functions by virtue of their unique dynamic instability processes which alternate between polymerization and depolymerization phases. This in turn is driven by a precise balance between attraction and repulsion forces between the constituents of microtubules (MTs)-tubulin dimers. Therefore, it is critically important to know what contributions result in a balance of the interaction energy among tubulin dimers that make up microtubules and what interactions may tip this balance toward or away from a stable polymerized state of tubulin. In this paper, we calculate the dipole-dipole interaction energy between tubulin dimers in a microtubule as part of the various contributions to the energy balance. We also compare the remaining contributions to the interaction energies between tubulin dimers and establish a balance between stabilizing and destabilizing components, including the van der Waals, electrostatic, and solvent-accessible surface area energies. The energy balance shows that the GTP-capped tip of the seam at the plus end of microtubules is stabilized only by - 9 kcal/mol, which can be completely reversed by the hydrolysis of a single GTP molecule, which releases + 14 kcal/mol and destabilizes the seam by an excess of + 5 kcal/mol. This triggers the breakdown of microtubules and initiates a disassembly phase which is aptly called a catastrophe.

Insights into the Effect of the G245S Single Point Mutation on the Structure of p53 and the Binding of the Protein to DNA.

The transcription factor p53 is a potent tumor suppressor dubbed as the "guardian of the genome" because of its ability to orchestrate protective biological outputs in response to a variety of oncogenic stresses. Mutation and thus inactivation of p53 can be found in 50% of human tumors. The majority are missense mutations located in the DNA binding region. Among them, G245S is known to be a structural hotspot mutation. To understand the behaviors and differences between the wild-type and mutant, both a dimer of the wild type p53 (wt-p53) and its G245S mutant (G245S-mp53), complexed with DNA, were simulated using molecular dynamics for more than 1 µs. wt-p53 and G245S-mp53 apo monomers were simulated for 1 µs as well. Conformational analyses and binding energy evaluations performed underline important differences and therefore provide insights to understand the G245S-mp53 loss of function. Our results indicate that the G245S mutation destabilizes several structural regions in the protein that are crucial for DNA binding when found in its apo form and highlight differences in the mutant-DNA complex structure compared to the wt protein. These findings not only provide means that can be applied to other p53 mutants but also serve as structural basis for further studies aimed at the development of cancer therapies based on restoring the function of p53.

Conformational fluctuations of the AXH monomer of Ataxin-1.

In this paper, we report the results of molecular dynamics simulations of AXH monomer of Ataxin-1. The AXH domain plays a crucial role in Ataxin-1 aggregation, which accompanies the initiation and progression of Spinocerebellar ataxia type 1. Our simulations involving both classical and replica exchange molecular dynamics, followed by principal component analysis of the trajectories obtained, reveal substantial conformational fluctuations of the protein structure, especially in the N-terminal region. We show that these fluctuations can be generated by thermal noise since the free energy barriers between conformations are small enough for thermally stimulated transitions. In agreement with the previous experimental findings, our results can be considered as a basis for a future design of ataxin aggregation inhibitors that will require several key conformations identified in the present study as molecular targets for ligand binding.

Characterization of the AXH domain of Ataxin-1 using enhanced sampling and functional mode analysis.

Ataxin-1 is the protein responsible for the Spinocerebellar ataxia type 1, an incurable neurodegenerative disease caused by polyglutamine expansion. The AXH domain plays a pivotal role in physiological functions of Ataxin-1. In Spinocerebellar ataxia 1, the AXH domain is involved in the misfolding and aggregation pathway. Here molecular modeling is applied to investigate the protein-protein interactions contributing to the AXH dimer stability. Particular attention is focused on: (i) the characterization of AXH monomer-monomer interface; (ii) the molecular description of the AXH monomer-monomer interaction dynamics. Technically, an approach based on functional mode analysis, here applied to replica exchange molecular dynamics trajectories, was employed. The findings of this study are consistent with previous experimental results and elucidate the pivotal role of the I580 residue in mediating the AXH monomer-monomer interaction dynamics.