Integrative workflows for the characterization of hydrophobin and cerato-platanin in the marine fungus Paradendryphiella salina.

Hydrophobins (HFBs) and cerato-platanins (CPs) are surface-active extracellular proteins produced by filamentous fungi. This study identified two HFB genes (pshyd1 and pshyd2) and one CP gene (pscp) in the marine fungus Paradendryphiella salina. The proteins PsCP, PsHYD2, and PsHYD1 had molecular weights of 12.70, 6.62, and 5.98 kDa, respectively, with isoelectric points below 7. PsHYD1 and PsHYD2 showed hydrophobicity (GRAVY score 0.462), while PsCP was hydrophilic (GRAVY score - 0.202). Stability indices indicated in-solution stability. Mass spectrometry identified 2,922 proteins, including CP but not HFB proteins. qPCR revealed differential gene expression influenced by developmental stage and substrate, with pshyd1 consistently expressed. These findings suggest P. salina's adaptation to marine ecosystems with fewer hydrophobin genes than other fungi but capable of producing surface-active proteins from seaweed carbohydrates. These proteins have potential applications in medical biocoatings, food industry foam stabilizers, and environmental bioremediation.

Protein Language Models and Machine Learning Facilitate the Identification of Antimicrobial Peptides.

Peptides are bioactive molecules whose functional versatility in living organisms has led to successful applications in diverse fields. In recent years, the amount of data describing peptide sequences and function collected in open repositories has substantially increased, allowing the application of more complex computational models to study the relations between the peptide composition and function. This work introduces AMP-Detector, a sequence-based classification model for the detection of peptides' functional biological activity, focusing on accelerating the discovery and de novo design of potential antimicrobial peptides (AMPs). AMP-Detector introduces a novel sequence-based pipeline to train binary classification models, integrating protein language models and machine learning algorithms. This pipeline produced 21 models targeting antimicrobial, antiviral, and antibacterial activity, achieving average precision exceeding 83%. Benchmark analyses revealed that our models outperformed existing methods for AMPs and delivered comparable results for other biological activity types. Utilizing the Peptide Atlas, we applied AMP-Detector to discover over 190,000 potential AMPs and demonstrated that it is an integrative approach with generative learning to aid in de novo design, resulting in over 500 novel AMPs. The combination of our methodology, robust models, and a generative design strategy offers a significant advancement in peptide-based drug discovery and represents a pivotal tool for therapeutic applications.

Early mutational signatures and transmissibility of SARS-CoV-2 Gamma and Lambda variants in Chile.

Genomic surveillance (GS) programmes were crucial in identifying and quantifying the mutating patterns of SARS-CoV-2 during the COVID-19 pandemic. In this work, we develop a Bayesian framework to quantify the relative transmissibility of different variants tailored for regions with limited GS. We use it to study the relative transmissibility of SARS-CoV-2 variants in Chile. Among the 3443 SARS-CoV-2 genomes collected between January and June 2021, where sampling was designed to be representative, the Gamma (P.1), Lambda (C.37), Alpha (B.1.1.7), B.1.1.348, and B.1.1 lineages were predominant. We found that Lambda and Gamma variants' reproduction numbers were 5% (95% CI: [1%, 14%]) and 16% (95% CI: [11%, 21%]) larger than Alpha's, respectively. Besides, we observed a systematic mutation enrichment in the Spike gene for all circulating variants, which strongly correlated with variants' transmissibility during the studied period (r = 0.93, p-value = 0.025). We also characterised the mutational signatures of local samples and their evolution over time and with the progress of vaccination, comparing them with those of samples collected in other regions worldwide. Altogether, our work provides a reliable method for quantifying variant transmissibility under subsampling and emphasises the importance of continuous genomic surveillance.

Peptide-based drug discovery through artificial intelligence: towards an autonomous design of therapeutic peptides.

With their diverse biological activities, peptides are promising candidates for therapeutic applications, showing antimicrobial, antitumour and hormonal signalling capabilities. Despite their advantages, therapeutic peptides face challenges such as short half-life, limited oral bioavailability and susceptibility to plasma degradation. The rise of computational tools and artificial intelligence (AI) in peptide research has spurred the development of advanced methodologies and databases that are pivotal in the exploration of these complex macromolecules. This perspective delves into integrating AI in peptide development, encompassing classifier methods, predictive systems and the avant-garde design facilitated by deep-generative models like generative adversarial networks and variational autoencoders. There are still challenges, such as the need for processing optimization and careful validation of predictive models. This work outlines traditional strategies for machine learning model construction and training techniques and proposes a comprehensive AI-assisted peptide design and validation pipeline. The evolving landscape of peptide design using AI is emphasized, showcasing the practicality of these methods in expediting the development and discovery of novel peptides within the context of peptide-based drug discovery.

The role of machine learning in health policies during the COVID-19 pandemic and in long COVID management.

The ongoing COVID-19 pandemic is arguably one of the most challenging health crises in modern times. The development of effective strategies to control the spread of SARS-CoV-2 were major goals for governments and policy makers. Mathematical modeling and machine learning emerged as potent tools to guide and optimize the different control measures. This review briefly summarizes the SARS-CoV-2 pandemic evolution during the first 3 years. It details the main public health challenges focusing on the contribution of mathematical modeling to design and guide government action plans and spread mitigation interventions of SARS-CoV-2. Next describes the application of machine learning methods in a series of study cases, including COVID-19 clinical diagnosis, the analysis of epidemiological variables, and drug discovery by protein engineering techniques. Lastly, it explores the use of machine learning tools for investigating long COVID, by identifying patterns and relationships of symptoms, predicting risk indicators, and enabling early evaluation of COVID-19 sequelae.

Generalized Property-Based Encoders and Digital Signal Processing Facilitate Predictive Tasks in Protein Engineering.

Computational methods in protein engineering often require encoding amino acid sequences, i.e., converting them into numeric arrays. Physicochemical properties are a typical choice to define encoders, where we replace each amino acid by its value for a given property. However, what property (or group thereof) is best for a given predictive task remains an open problem. In this work, we generalize property-based encoding strategies to maximize the performance of predictive models in protein engineering. First, combining text mining and unsupervised learning, we partitioned the AAIndex database into eight semantically-consistent groups of properties. We then applied a non-linear PCA within each group to define a single encoder to represent it. Then, in several case studies, we assess the performance of predictive models for protein and peptide function, folding, and biological activity, trained using the proposed encoders and classical methods (One Hot Encoder and TAPE embeddings). Models trained on datasets encoded with our encoders and converted to signals through the Fast Fourier Transform (FFT) increased their precision and reduced their overfitting substantially, outperforming classical approaches in most cases. Finally, we propose a preliminary methodology to create de novo sequences with desired properties. All these results offer simple ways to increase the performance of general and complex predictive tasks in protein engineering without increasing their complexity.

Patient-Wise Methodology to Assess Glycemic Health Status: Applications to Quantify the Efficacy and Physiological Targets of Polyphenols on Glycemic Control.

A growing body of evidence indicates that dietary polyphenols could be used as an early intervention to treat glucose-insulin (G-I) dysregulation. However, studies report heterogeneous information, and the targets of the intervention remain largely elusive. In this work, we provide a general methodology to quantify the effects of any given polyphenol-rich food or formulae over glycemic regulation in a patient-wise manner using an Oral Glucose Tolerance Test (OGTT). We use a mathematical model to represent individual OGTT curves as the coordinated action of subsystems, each one described by a parameter with physiological interpretation. Using the parameter values calculated for a cohort of 1198 individuals, we propose a statistical model to calculate the risk of dysglycemia and the coordination among subsystems for each subject, thus providing a continuous and individual health assessment. This method allows identifying individuals at high risk of dysglycemia-which would have been missed with traditional binary diagnostic methods-enabling early nutritional intervention with a polyphenol-supplemented diet where it is most effective and desirable. Besides, the proposed methodology assesses the effectiveness of interventions over time when applied to the OGTT curves of a treated individual. We illustrate the use of this method in a case study to assess the dose-dependent effects of Delphinol® on reducing dysglycemia risk and improving the coordination between subsystems. Finally, this strategy enables, on the one hand, the use of low-cost, non-invasive methods in population-scale nutritional studies. On the other hand, it will help practitioners assess the effectiveness of an intervention based on individual vulnerabilities and adapt the treatment to manage dysglycemia and avoid its progression into disease.

Peptipedia: a user-friendly web application and a comprehensive database for peptide research supported by Machine Learning approach.

Peptides have attracted attention during the last decades due to their extraordinary therapeutic properties. Different computational tools have been developed to take advantage of existing information, compiling knowledge and making available the information for common users. Nevertheless, most related tools available are not user-friendly, present redundant information, do not clearly display the data, and usually are specific for particular biological activities, not existing so far, an integrated database with consolidated information to help research peptide sequences. To solve these necessities, we developed Peptipedia, a user-friendly web application and comprehensive database to search, characterize and analyse peptide sequences. Our tool integrates the information from 30 previously reported databases with a total of 92 055 amino acid sequences, making it the biggest repository of peptides with recorded activities to date. Furthermore, we make available a variety of bioinformatics services and statistical modules to increase our tool's usability. Moreover, we incorporated a robust assembled binary classification system to predict putative biological activities for peptide sequences. Our tools' significant differences with other existing alternatives become a substantial contribution for developing biotechnological and bioengineering applications for peptides. Peptipedia is available for non-commercial use as an open-access software, licensed under the GNU General Public License, version GPL 3.0. The web platform is publicly available at peptipedia.cl. Database URL: Both the source code and sample data sets are available in the GitHub repository https://github.com/ProteinEngineering-PESB2/peptipedia.

Statistically-based methodology for revealing real contagion trends and correcting delay-induced errors in the assessment of COVID-19 pandemic.

COVID-19 pandemic has reshaped our world in a timescale much shorter than what we can understand. Particularities of SARS-CoV-2, such as its persistence in surfaces and the lack of a curative treatment or vaccine against COVID-19, have pushed authorities to apply restrictive policies to control its spreading. As data drove most of the decisions made in this global contingency, their quality is a critical variable for decision-making actors, and therefore should be carefully curated. In this work, we analyze the sources of error in typically reported epidemiological variables and usual tests used for diagnosis, and their impact on our understanding of COVID-19 spreading dynamics. We address the existence of different delays in the report of new cases, induced by the incubation time of the virus and testing-diagnosis time gaps, and other error sources related to the sensitivity/specificity of the tests used to diagnose COVID-19. Using a statistically-based algorithm, we perform a temporal reclassification of cases to avoid delay-induced errors, building up new epidemiologic curves centered in the day where the contagion effectively occurred. We also statistically enhance the robustness behind the discharge/recovery clinical criteria in the absence of a direct test, which is typically the case of non-first world countries, where the limited testing capabilities are fully dedicated to the evaluation of new cases. Finally, we applied our methodology to assess the evolution of the pandemic in Chile through the Effective Reproduction Number Rt , identifying different moments in which data was misleading governmental actions. In doing so, we aim to raise public awareness of the need for proper data reporting and processing protocols for epidemiological modelling and predictions.

A multi-group SEIRA model for the spread of COVID-19 among heterogeneous populations.

The outbreak and propagation of COVID-19 have posed a considerable challenge to modern society. In particular, the different restrictive actions taken by governments to prevent the spread of the virus have changed the way humans interact and conceive interaction. Due to geographical, behavioral, or economic factors, different sub-groups among a population are more (or less) likely to interact, and thus to spread/acquire the virus. In this work, we present a general multi-group SEIRA model for representing the spread of COVID-19 among a heterogeneous population and test it in a numerical case of study. By highlighting its applicability and the ease with which its general formulation can be adapted to particular studies, we expect our model to lead us to a better understanding of the evolution of this pandemic and to better public-health policies to control it.

A Novel Synthetic Model of the Glucose-Insulin System for Patient-Wise Inference of Physiological Parameters From Small-Size OGTT Data.

Existing mathematical models for the glucose-insulin (G-I) dynamics often involve variables that are not susceptible to direct measurement. Standard clinical tests for measuring G-I levels for diagnosing potential diseases are simple and relatively cheap, but seldom give enough information to allow the identification of model parameters within the range in which they have a biological meaning, thus generating a gap between mathematical modeling and any possible physiological explanation or clinical interpretation. In the present work, we present a synthetic mathematical model to represent the G-I dynamics in an Oral Glucose Tolerance Test (OGTT), which involves for the first time for OGTT-related models, Delay Differential Equations. Our model can represent the radically different behaviors observed in a studied cohort of 407 normoglycemic patients (the largest analyzed so far in parameter fitting experiments), all masked under the current threshold-based normality criteria. We also propose a novel approach to solve the parameter fitting inverse problem, involving the clustering of different G-I profiles, a simulation-based exploration of the feasible set, and the construction of an information function which reshapes it, based on the clinical records, experimental uncertainties, and physiological criteria. This method allowed an individual-wise recognition of the parameters of our model using small size OGTT data (5 measurements) directly, without modifying the routine procedures or requiring particular clinical setups. Therefore, our methodology can be easily applied to gain parametric insights to complement the existing tools for the diagnosis of G-I dysregulations. We tested the parameter stability and sensitivity for individual subjects, and an empirical relationship between such indexes and curve shapes was spotted. Since different G-I profiles, under the light of our model, are related to different physiological mechanisms, the present method offers a tool for personally-oriented diagnosis and treatment and to better define new health criteria.

Development of Supervised Learning Predictive Models for Highly Non-linear Biological, Biomedical, and General Datasets.

In highly non-linear datasets, attributes or features do not allow readily finding visual patterns for identifying common underlying behaviors. Therefore, it is not possible to achieve classification or regression using linear or mildly non-linear hyperspace partition functions. Hence, supervised learning models based on the application of most existing algorithms are limited, and their performance metrics are low. Linear transformations of variables, such as principal components analysis, cannot avoid the problem, and even models based on artificial neural networks and deep learning are unable to improve the metrics. Sometimes, even when features allow classification or regression in reported cases, performance metrics of supervised learning algorithms remain unsatisfyingly low. This problem is recurrent in many areas of study as, per example, the clinical, biotechnological, and protein engineering areas, where many of the attributes are correlated in an unknown and very non-linear fashion or are categorical and difficult to relate to a target response variable. In such areas, being able to create predictive models would dramatically impact the quality of their outcomes, generating an immediate added value for both the scientific and general public. In this manuscript, we present RV-Clustering, a library of unsupervised learning algorithms, and a new methodology designed to find optimum partitions within highly non-linear datasets that allow deconvoluting variables and notoriously improving performance metrics in supervised learning classification or regression models. The partitions obtained are statistically cross-validated, ensuring correct representativity and no over-fitting. We have successfully tested RV-Clustering in several highly non-linear datasets with different origins. The approach herein proposed has generated classification and regression models with high-performance metrics, which further supports its ability to generate predictive models for highly non-linear datasets. Advantageously, the method does not require significant human input, which guarantees a higher usability in the biological, biomedical, and protein engineering community with no specific knowledge in the machine learning area.

Real-Time Estimation of R t for Supporting Public-Health Policies Against COVID-19.

In the absence of a consensus protocol to slow down the spread of SARS-CoV-2, policymakers need real-time indicators to support decisions in public health matters. The Effective Reproduction Number (R t ) represents the number of secondary infections generated per each case and can be dramatically modified by applying effective interventions. However, current methodologies to calculate R t from data remain somewhat cumbersome, thus raising a barrier between its timely calculation and application by policymakers. In this work, we provide a simple mathematical formulation for obtaining the effective reproduction number in real-time using only and directly daily official case reports, obtained by modifying the equations describing the viral spread. We numerically explore the accuracy and limitations of the proposed methodology, which was demonstrated to provide accurate, timely, and intuitive results. We illustrate the use of our methodology to study the evolution of the pandemic in different iconic countries, and to assess the efficacy and promptness of different public health interventions.