Combined mechanistic modeling and machine-learning approaches in systems biology - A systematic literature review

Evolution of gene regulatory networks in insects

Changes in gene regulatory networks (GRNs) underlying the evolution of traits have been intensively studied, with insects providing excellent model cases. In studies using Drosophila, butterflies, and other insects, several well-known cases have shown that changes in the cis-regulatory region of a gene controlling a trait can result in the co-option of the gene for a role different from that in its original developmental context. When the expression of a regulatory gene that controls the expression of multiple downstream genes is altered, the expression of these downstream genes changes accordingly, representing the simplest form of GRN co-option. Many studies have explored the applicability of this model to the acquisition of new traits, yielding substantial insights. However, no study has yet comprehensively elucidated the co-option of a GRN or the evolution of a network architecture, including associated genes and their regulatory relationships. In the near future, the use of single-cell multiomics and machine learning will allow for larger-scale data analysis, leading to a better understanding of the evolution of traits through the evolution of GRNs.

A Python toolbox for neural circuit parameter inference

Computational research tools have reached a level of maturity that enables efficient simulation of neural activity across diverse scales. Concurrently, experimental neuroscience is experiencing an unprecedented scale of data generation. Despite these advancements, our understanding of the precise mechanistic relationship between neural recordings and key aspects of neural activity remains insufficient, including which specific features of electrophysiological population dynamics (i.e., putative biomarkers) best reflect properties of the underlying microcircuit configuration. We present ncpi, an open-source Python toolbox that serves as an all-in-one solution, effectively integrating well-established methods for both forward and inverse modeling of extracellular signals based on single-neuron network model simulations. Our tool serves as a benchmarking resource for model-driven interpretation of electrophysiological data and the evaluation of candidate biomarkers that plausibly index changes in neural circuit parameters. Using mouse LFP data and human EEG recordings, we demonstrate the potential of ncpi to uncover imbalances in neural circuit parameters during brain development and in Alzheimer's Disease.

Artificial-intelligence-driven Innovations in Mechanistic Computational Modeling and Digital Twins for Biomedical Applications

Understanding of complex biological systems remains a significant challenge due to their high dimensionality, nonlinearity, and context-specific behavior. Artificial intelligence (AI) and mechanistic modeling are becoming essential tools for studying such complex systems. Mechanistic modeling can facilitate the construction of simulatable models that are interpretable but often struggle with scalability and parameters estimation. AI can integrate multi-omics data to create predictive models, but it lacks interpretability. The gap between these two modeling methods limits our ability to develop comprehensive and predictive models for biomedical applications. This article reviews the most recent advancements in the integration of AI and mechanistic modeling to fill this gap. Recently, with omics availability, AI has led to new discoveries in mechanistic computational modeling. The mechanistic models can also help in getting insight into the mechanism for prediction made by AI models. This integration is helpful in modeling complex systems, estimating the parameters that are hard to capture in experiments, and creating surrogate models to reduce computational costs because of expensive mechanistic model simulations. This article focuses on advancements in mechanistic computational models and AI models and their integration for scientific discoveries in biology, pharmacology, drug discovery and diseases. The mechanistic models with AI integration can facilitate biological discoveries to advance our understanding of disease mechanisms, drug development, and personalized medicine. The article also highlights the role of AI and mechanistic model integration in the development of more advanced models in the biomedical domain, such as medical digital twins and virtual patients for pharmacological discoveries.

Advancing in Silico Clinical Trials for Regulatory Adoption and Innovation

The evolution of information and communication technologies has affected all fields of science, including health sciences. However, the rate of technological innovation adoption by the healthcare sector has been historically slow, compared to other industrial sectors. Innovation in computer modeling and simulation approaches has changed the landscape in biomedical applications and biomedicine, paving the way for their potential contribution in reducing, refining, and partially replacing animal and human clinical trials. In Silico Clinical Trials (ISCT) allow the development of virtual populations used in the safety and efficacy testing of new drugs and medical devices. This White Paper presents the current framework for ISCT, the role of in silico medicine research communities, the different perspectives (research, scientific, clinical, regulatory, standardization, data quality, legal and ethical), the barriers, challenges, and opportunities for ISCT adoption. In addition, an overview of successful ISCT projects, market-available platforms, and FDA- approved paradigms, along with their vision, mission and outcomes are presented.

Simultaneous Multi-Treatment Strategy for Brain Tumor Reduction via Nonlinear Control

Background: Recently proposed brain-tumor treatment strategies prioritize fast reduction of tumor cell population while often neglecting the radiation or chemotherapeutic drug dosage requirements to achieve it. Moreover, these techniques provide chemotherapy based treatment strategies, while ignoring the toxic side effects of the drugs employed by it. Methods: This study updates the recently proposed brain-tumor system dynamics by incorporating radiotherapy along with chemotherapy to simultaneously initiate both therapies for a more comprehensive and effective response against tumor proliferation. Afterwards, based on the upgraded system dynamics, this study proposes a novel multi-input sigmoid-based smooth synergetic nonlinear controller with the aim to reduce the dosage requirements of both therapies while keeping the overall system response robust and efficient. The novelty of this study lies in the combination of radiotherapy and chemotherapy inputs in a way that prioritizes patients health and well-being, while integrating advanced synergetic control technique with a sigmoid function based smoothing agent. Results: The proposed method reduced baseline radiation and chemo drug dosages by 57% and 33% respectively while effectively suppressing tumor growth and proliferation. Similarly, the proposed controller reduced the time required for complete tumor mitigation by 60% while reducing the radiation and chemotherapeutic drug intensity by 93.8% and 21.3% respectively. Conclusions: This study offers significant improvement in tumor treatment methodologies by providing a safer, less riskier brain-tumor treatment strategy that has promising potential to improve survival rates against this menacing health condition so that the affected patients may lead a healthier and better quality of life.

Deep-learning-ready RGB-depth images of seedling development

In the era of machine learning-driven plant imaging, the production of annotated datasets is a very important contribution. In this data paper, a unique annotated dataset of seedling emergence kinetics is proposed. It is composed of almost 70,000 RGB-depth frames and more than 700,000 plant annotations. The dataset is shown valuable for training deep learning models and performing high-throughput phenotyping by imaging. The ability of such models to generalize to several species and outperform the state-of-the-art owing to the delivered dataset is demonstrated. We also discuss how this dataset raises new questions in plant phenotyping.

Investigating and forecasting infectious disease dynamics using epidemiological and molecular surveillance data

The integration of viral genomic data into public health surveillance has revolutionized our ability to track and forecast infectious disease dynamics. This review addresses two critical aspects of infectious disease forecasting and monitoring: the methodological workflow for epidemic forecasting and the transformative role of molecular surveillance. We first present a detailed approach for validating epidemic models, emphasizing an iterative workflow that utilizes ordinary differential equation (ODE)-based models to investigate and forecast disease dynamics. We recommend a more structured approach to model validation, systematically addressing key stages such as model calibration, assessment of structural and practical parameter identifiability, and effective uncertainty propagation in forecasts. Furthermore, we underscore the importance of incorporating multiple data streams by applying both simulated and real epidemiological data from the COVID-19 pandemic to produce more reliable forecasts with quantified uncertainty. Additionally, we emphasize the pivotal role of viral genomic data in tracking transmission dynamics and pathogen evolution. By leveraging advanced computational tools such as Bayesian phylogenetics and phylodynamics, researchers can more accurately estimate transmission clusters and reconstruct outbreak histories, thereby improving data-driven modeling and forecasting and informing targeted public health interventions. Finally, we discuss the transformative potential of integrating molecular epidemiology with mathematical modeling to complement and enhance epidemic forecasting and optimize public health strategies.

Enhancing drug-target interaction predictions in context of neurodegenerative diseases using bidirectional long short-term memory in male Swiss albino mice pharmaco-EEG analysis

Background and Objective

Emerging diseases like Parkinson or Alzheimer's, which are not curable, endanger human mental health and are challenging to research. Drug target interactions (DTI) are pivotal in the screening of candidate drugs and focus on a small pool of drug targets. Electroencephalogram shows the responses to psychotropic medicines in the brain bioelectric activity. Synaptic activity can be analyzed by using Local Field Potential recordings obtained from micro-electrodes implanted in the brain. The aim is to evaluate the effects of drug on brain bioelectric activity and increase the drug classification accuracy. The ultimate goal is to advance our understanding of how drugs affect synaptic activity and open the door to more focused treatment for neurodegenerative diseases.

Methods

In this study, Pharmaco-EEG recordings are processed using Advanced neural network models, particularly Convolutional Neural Networks, to assess the effects of medications. The five different medicines used in this study are Ephedrine, Fluoxetine, Kratom, Morphine, and Saline. The signals observed are local field potential signals. To overcome some limits of DTI prediction, we propose Bidirectional Long Short-Term Memory (LSTM) for the categorization of intracranial EEG (i-EEG) data, departing from standard approaches. Similar EEG patterns are presumably caused by drugs that work by homologous pharmacological pathways, producing similar psychotropic effects. To improve accuracy and reduce training loss, our study introduces a bidirectional LSTM model for classification along with Bayesian optimization.

Results

High recall, precision, and F1-Scores, particularly a 95% F1-Score for morphine, ephedrine, fluoxetine, and saline, suggest good performance in predicting these drug classes. Kratom produces a somewhat lower recall of 94%, but a high F1-Score of 97% and perfect precision of 1.00. The weighted average F1-Score, macro average, and overall accuracy are all consistently high (around 97%), indicating that the model works well throughout the spectrum of drugs.

Conclusions

Improved model performance was demonstrated by using a diversified dataset with five drug categories and bidirectional LSTM boosted with Bayesian optimization for hyperparameter tuning. From earlier limited-category models, it represents a substantial advancement.

Investigation of the link between first-order kinetic models of the in vitro digestion of native starches and the accompanying changes in their crystallinity and structure

Starch is the main source of dietary energy for humans. In order to understand the mechanisms governing native starch in vitro digestion, digestion data for six starches [wheat, maize, (waxy) maize, rice, potato and pea] of different botanical sources were fitted with the most common first-order kinetic models, i.e. the single, sequential, parallel and combined models. Parallel and combined models provided the most accurate fits and showed that all starches studied except potato starch followed a biphasic in vitro digestion pattern. The biological relevance of the kinetic parameters was explored by determining changes in crystallinity and molecular structure of the undigested starch residues during in vitro digestion. While the crystallinity of the undigested potato starch residues did not change substantially, a respectively small and large decrease in their amylose content and chain length during in vitro digestion was observed, indicating that amylose was digested slightly preferentially over amylopectin in native starch. However, the molecular structure of the starch residues changed too slowly and/or only to an insufficient extent to relate it to the kinetic parameters of the digested fractions predicted by the models. Such parameters thus need to be interpreted with caution, as their biological relevance still needs to be proven.

Modelling the age pattern of fertility: an individual-level approach

Fitting statistical models to aggregate data is still the dominant approach in many demographic and biodemographic applications. Although these macro-level models have proven useful for a variety of tasks, they often have no demographic interpretation. Individual-level modelling, on the other hand, offers a deeper understanding of the mechanisms underlying observed patterns. Their parameters represent quantities in the real world, instead of pure mathematical abstractions. However, estimating these parameters using real-world data has remained a challenge. The approach we introduce in this article attempts to overcome this limitation. Using a likelihood-free inference technique, we show that it is possible to estimate the parameters of a simple but demographically interpretable individual-level model of the reproductive process by exclusively relying on the information contained in a set of age-specific fertility rates. By estimating individual-level models from widely available aggregate data, this approach can contribute to a better understanding of reproductive behaviour and its driving mechanisms, bridging the gap between individual-level and population-level processes. We illustrate our approach using data from three natural fertility populations.

Physiological model-based machine learning for classifying patients with binge-eating disorder (BED) from the Oral Glucose Tolerance Test (OGTT) curve

BACKGROUND AND OBJECTIVE

Binge eating disorder (BED) is the most frequent eating disorder, often confused with obesity, with which it shares several characteristics. Early identification could enable targeted therapeutic interventions. In this study, we propose a hybrid pipeline that, starting from plasma glucose data acquired during the Oral Glucose Tolerance Test (OGTT), allows us to classify the two types of patients through computational modeling and artificial intelligence.

METHODS

The proposed hybrid pipeline integrates a classical mechanistic model of delayed differential equations (DDE) that describes glucose-insulin dynamics with machine learning (ML) methods. Ad hoc techniques, including structural identifiability analysis, have been employed for refining and evaluating the mathematical model. Additionally, a dedicated pipeline for identifying and optimizing model parameters has been applied to obtain reliable estimates. Robust feature extraction and classifier selection processes were developed to ensure the optimal choice of the best-performing classifier.

RESULTS

By leveraging parameters estimated from the mechanistic model alongside easily obtainable patient information (such as glucose levels at 30 and 120 min post-OGTT, glycated hemoglobin (Hb1Ac), body mass index (BMI), and waist circumference), our approach facilitates accurate classification of patients, enabling tailored therapeutic interventions.

CONCLUSION

Initial findings, focusing on correctly categorizing patients with BED based on metabolic data, demonstrate promising outcomes. These results suggest significant potential for refinement, including exploration of alternative mechanistic models and machine learning algorithms, to enhance classification accuracy and therapeutic strategies.

Artificial Intelligence for Climate Change Biology: From Data Collection to Predictions

In the era of big data, ecological research is experiencing a transformative shift, yet big-data advancements in thermal ecology and the study of animal responses to climate conditions remain limited. This review discusses how big data analytics and artificial intelligence (AI) can significantly enhance our understanding of microclimates and animal behaviors under changing climatic conditions. We explore AI's potential to refine microclimate models and analyze data from advanced sensors and camera technologies, which capture detailed, high-resolution information. This integration can allow researchers to dissect complex ecological and physiological processes with unprecedented precision. We describe how AI can enhance microclimate modeling through improved bias correction and downscaling techniques, providing more accurate estimates of the conditions that animals face under various climate scenarios. Additionally, we explore AI's capabilities in tracking animal responses to these conditions, particularly through innovative classification models that utilize sensors such as accelerometers and acoustic loggers. For example, the widespread usage of camera traps can benefit from AI-driven image classification models to accurately identify thermoregulatory responses, such as shade usage and panting. AI is therefore instrumental in monitoring how animals interact with their environments, offering vital insights into their adaptive behaviors. Finally, we discuss how these advanced data-driven approaches can inform and enhance conservation strategies. In particular, detailed mapping of microhabitats essential for species survival under adverse conditions can guide the design of climate-resilient conservation and restoration programs that prioritize habitat features crucial for biodiversity resilience. In conclusion, the convergence of AI, big data, and ecological science heralds a new era of precision conservation, essential for addressing the global environmental challenges of the 21st century.

Machine Learning for Biomedical Applications

Machine learning (ML) is a field of artificial intelligence that uses algorithms capable of extracting knowledge directly from data that could support decisions in multiple fields of engineering [...].

Postmortem metabolomics: influence of time since death on the level of endogenous compounds in human femoral blood. Necessary to be considered in metabolome study planning?

INTRODUCTION

The (un)targeted analysis of endogenous compounds has gained interest in the field of forensic postmortem investigations. The blood metabolome is influenced by many factors, and postmortem specimens are considered particularly challenging due to unpredictable decomposition processes.

OBJECTIVES

This study aimed to systematically investigate the influence of the time since death on endogenous compounds and its relevance in designing postmortem metabolome studies.

METHODS

Femoral blood samples of 427 authentic postmortem cases, were collected at two time points after death (854 samples in total; t1: admission to the institute, 1.3-290 h; t2: autopsy, 11-478 h; median ∆t = 71 h). All samples were analyzed using an untargeted metabolome approach, and peak areas were determined for 38 compounds (acylcarnitines, amino acids, phospholipids, and others). Differences between t2 and t1 were assessed by Wilcoxon signed-ranked test (p < 0.05). Moreover, all samples (n = 854) were binned into time groups (6 h, 12 h, or 24 h intervals) and compared by Kruskal-Wallis/Dunn's multiple comparison tests (p < 0.05 each) to investigate the effect of the estimated time since death.

RESULTS

Except for serine, threonine, and PC 34:1, all tested analytes revealed statistically significant changes between t1 and t2 (highest median increase 166%). Unpaired analysis of all 854 blood samples in-between groups indicated similar results. Significant differences were typically observed between blood samples collected within the first and later than 48 h after death, respectively.

CONCLUSIONS

To improve the consistency of comprehensive data evaluation in postmortem metabolome studies, it seems advisable to only include specimens collected within the first 2 days after death.

A hybrid modeling framework for generalizable and interpretable predictions of ICU mortality across multiple hospitals

The development of reliable mortality risk stratification models is an active research area in computational healthcare. Mortality risk stratification provides a standard to assist physicians in evaluating a patient's condition or prognosis objectively. Particular interest lies in methods that are transparent to clinical interpretation and that retain predictive power once validated across diverse datasets they were not trained on. This study addresses the challenge of consolidating numerous ICD codes for predictive modeling of ICU mortality, employing a hybrid modeling approach that integrates mechanistic, clinical knowledge with mathematical and machine learning models . A tree-structured network connecting independent modules that carry clinical meaning is implemented for interpretability. Our training strategy utilizes graph-theoretic methods for data analysis, aiming to identify the functions of individual black-box modules within the tree-structured network by harnessing solutions from specific max-cut problems. The trained model is then validated on external datasets from different hospitals, demonstrating successful generalization capabilities, particularly in binary-feature datasets where label assessment involves extrapolation.

Combining simulation models and machine learning in healthcare management: strategies and applications

Simulation models and artificial intelligence (AI) are largely used to address healthcare and biomedical engineering problems. Both approaches showed promising results in the analysis and optimization of healthcare processes. Therefore, the combination of simulation models and AI could provide a strategy to further boost the quality of health services. In this work, a systematic review of studies applying a hybrid simulation models and AI approach to address healthcare management challenges was carried out. Scopus, Web of Science, and PubMed databases were screened by independent reviewers. The main strategies to combine simulation and AI as well as the major healthcare application scenarios were identified and discussed. Moreover, tools and algorithms to implement the proposed approaches were described. Results showed that machine learning appears to be the most employed AI strategy in combination with simulation models, which mainly rely on agent-based and discrete-event systems. The scarcity and heterogeneity of the included studies suggested that a standardized framework to implement hybrid machine learning-simulation approaches in healthcare management is yet to be defined. Future efforts should aim to use these approaches to design novel intelligentin-silicomodels of healthcare processes and to provide effective translation to the clinics.

Mechanism-aware and multimodal AI: beyond model-agnostic interpretation

Artificial intelligence (AI) is widely used for exploiting multimodal biomedical data, with increasingly accurate predictions and model-agnostic interpretations, which are however also agnostic to biological mechanisms. Combining metabolic modelling, 'omics, and imaging data via multimodal AI can generate predictions that can be interpreted mechanistically and transparently, therefore with significantly higher therapeutic potential.

How Can Proteomics Help to Elucidate the Pathophysiological Crosstalk in Muscular Dystrophy and Associated Multi-System Dysfunction?

This perspective article is concerned with the question of how proteomics, which is a core technique of systems biology that is deeply embedded in the multi-omics field of modern bioresearch, can help us better understand the molecular pathogenesis of complex diseases. As an illustrative example of a monogenetic disorder that primarily affects the neuromuscular system but is characterized by a plethora of multi-system pathophysiological alterations, the muscle-wasting disease Duchenne muscular dystrophy was examined. Recent achievements in the field of dystrophinopathy research are described with special reference to the proteome-wide complexity of neuromuscular changes and body-wide alterations/adaptations. Based on a description of the current applications of top-down versus bottom-up proteomic approaches and their technical challenges, future systems biological approaches are outlined. The envisaged holistic and integromic bioanalysis would encompass the integration of diverse omics-type studies including inter- and intra-proteomics as the core disciplines for systematic protein evaluations, with sophisticated biomolecular analyses, including physiology, molecular biology, biochemistry and histochemistry. Integrated proteomic findings promise to be instrumental in improving our detailed knowledge of pathogenic mechanisms and multi-system dysfunction, widening the available biomarker signature of dystrophinopathy for improved diagnostic/prognostic procedures, and advancing the identification of novel therapeutic targets to treat Duchenne muscular dystrophy.

Functional selectivity of Receptor Tyrosine Kinases regulates distinct cellular outputs

Functional selectivity refers to the activation of differential signalling and cellular outputs downstream of the same membrane-bound receptor when activated by two or more different ligands. Functional selectivity has been described and extensively studied for G-protein Coupled Receptors (GPCRs), leading to specific therapeutic options for dysregulated GPCRs functions. However, studies regarding the functional selectivity of Receptor Tyrosine Kinases (RTKs) remain sparse. Here, we will summarize recent data about RTK functional selectivity focusing on how the nature and the amount of RTK ligands and the crosstalk of RTKs with other membrane proteins regulate the specificity of RTK signalling. In addition, we will discuss how structural changes in RTKs upon ligand binding affects selective signalling pathways. Much remains to be known about the integration of different signals affecting RTK signalling specificity to orchestrate long-term cellular outcomes. Recent advancements in omics, specifically quantitative phosphoproteomics, and in systems biology methods to study, model and integrate different types of large-scale omics data have increased our ability to compare several signals affecting RTK functional selectivity in a global, system-wide fashion. We will discuss how such methods facilitate the exploration of important signalling hubs and enable data-driven predictions aiming at improving the efficacy of therapeutics for diseases like cancer, where redundant RTK signalling pathways often compromise treatment efficacy.