Machine Learning for Biological Design.

Classification and prediction tasks are common in health research. With the increasing availability of vast health data repositories (e.g. electronic medical record databases) and advances in computing power, traditional statistical approaches are being augmented or replaced with machine learning (ML) approaches to classify and predict health outcomes. ML describes the automated process of identifying ("learning") patterns in data to perform tasks. Developing an ML model includes selecting between many ML models (e.g. decision trees, support vector machines, neural networks); model specifications such as hyperparameter tuning; and evaluation of model performance. This process is conducted repeatedly to find the model and corresponding specifications that optimize some measure of model performance. ML models can make more accurate classifications and predictions than their statistical counterparts and confer greater flexibility when modelling unstructured data or interactions between covariates; however, many ML models require larger sample sizes to achieve good classification or predictive performance and have been criticized as "black box" for their poor transparency and interpretability. ML holds potential in family medicine for risk profiling of patients' disease risk and clinical decision support to present additional information at times of uncertainty or high demand. In the future, ML approaches are positioned to become commonplace in family medicine. As such, it is important to understand the objectives that can be addressed using ML approaches and the associated techniques and limitations. This article provides a brief introduction into the use of ML approaches for classification and prediction tasks in family medicine.

Introduction Spine trauma represents a major cause of disability and substantial healthcare costs. Despite classification systems and risk scores, outcome prediction following spine trauma remains challenging. Machine learning (ML) approaches offer multidimensionality but vary widely in methodology, performance, and clinical application. This systematic review aims to characterize the use of ML for outcome prediction and prognostication in spine trauma. Methods A systematic search of PubMed, Embase, Scopus, Web of Science, and Cochrane Library was performed for publications from January 2015 to May 2025. Studies reporting ML for predicting clinical outcomes in spine trauma patients were identified using pre-defined criteria. Primary outcomes were types and performance metrics of ML models for spine trauma prognostication. Secondary outcomes included cost-effectiveness, clinical utility and implementation feasibility. Results Twenty-seven studies were included, of which 26 were published from 2020 onwards. Only one study was prospective, while the rest were retrospective. Commonly used ML algorithms for spine trauma prognosis were logistic regression (LR), extreme gradient boosting (XGBoost), Random Forest (RF), light gradient boosting machine (LightGBM), and support vector machine (SVM). Outcomes predicted include mortality, neurological function or recovery, ambulation and walking, self-care and independence, complications, treatment failure, length of stay, and discharge destination. Most studies employed internal testing and validation; only four reported external validation. Seven studies reported direct comparison with non-ML models or risk scores, demonstrating comparable or superior performance of ML models. Performance metrics include area under the curve (AUC), accuracy, sensitivity, specificity, precision, F1-score, Brier score, and others. The highest AUC was reported by a nomogram constructed using consensus clustering to predict postoperative pulmonary infection in patients with acute cervical spinal cord injury (internal validation; AUC 0.993; 95% CI 0.981–1.000). Quality of evidence varied between studies, but risk of bias was a concern. Conclusions Existing evidence on ML for spine trauma prognosis is characterized by retrospective study designs, homogeneous cohorts, and lack of external validation, limiting their generalizability and real-world application. While ML approaches demonstrate encouraging predictive performance for spine trauma outcomes, their clinical adoption is limited by methodological heterogeneity. More robust validation, standardized reporting, and comparison with traditional tools are needed before ML models can reliably inform clinical decision-making.

ABSTRACTIn human immunodeficiency virus (HIV) infection, virus replication in and adaptation to the central nervous system (CNS) can result in neurocognitive deficits in approximately 25% of patients with unsuppressed viremia. While no single viral mutation can be agreed upon as distinguishing the neuroadapted population, earlier studies have demonstrated that a machine learning (ML) approach could be applied to identify a collection of mutational signatures within the virus envelope glycoprotein (Gp120) predictive of disease. The S[imian]IV-infected macaque is a widely used animal model of HIV neuropathology, allowing in-depth tissue sampling infeasible for human patients. Yet, translational impact of the ML approach within the context of the macaque model has not been tested, much less the capacity for early prediction in other, noninvasive tissues. We applied the previously described ML approach to prediction of SIV-mediated encephalitis (SIVE) using gp120 sequences obtained from the CNS of animals with and without SIVE with 97% accuracy. The presence of SIVE signatures at earlier time points of infection in non-CNS tissues indicated these signatures cannot be used in a clinical setting; however, combined with protein structural mapping and statistical phylogenetic inference, results revealed common denominators associated with these signatures, including 2-acetamido-2-deoxy-beta-d-glucopyranose structural interactions and high rate of alveolar macrophage (AM) infection. AMs were also determined to be the phyloanatomic source of cranial virus in SIVE animals, but not in animals that did not develop SIVE, implicating a role for these cells in the evolution of the signatures identified as predictive of both HIV and SIV neuropathology.IMPORTANCE HIV-associated neurocognitive disorders remain prevalent among persons living with HIV (PLWH) owing to our limited understanding of the contributing viral mechanisms and ability to predict disease onset. We have expanded on a machine learning method previously used on HIV genetic sequence data to predict neurocognitive impairment in PLWH to the more extensively sampled SIV-infected macaque model in order to (i) determine the translatability of the animal model and (ii) more accurately characterize the predictive capacity of the method. We identified eight amino acid and/or biochemical signatures in the SIV envelope glycoprotein, the most predominant of which demonstrated the potential for aminoglycan interaction characteristic of previously identified HIV signatures. These signatures were not isolated to specific points in time or to the central nervous system, limiting their use as an accurate clinical predictor of neuropathogenesis; however, statistical phylogenetic and signature pattern analyses implicate the lungs as a key player in the emergence of neuroadapted viruses.

Active semi-supervised learning with multiple complementary information

Commentary: Machine learning and cardiac surgery risk prediction

The need for efficient computational screening of molecular candidates that possess desired properties frequently arises in various scientific and engineering problems, including drug discovery and materials design. However, the large size of the search space containing the candidates and the substantial computational cost of high-fidelity property prediction models makes screening practically challenging. In this work, we propose a general framework for constructing and optimizing a virtual screening (HTVS) pipeline that consists of multi-fidelity models. The central idea is to optimally allocate the computational resources to models with varying costs and accuracy to optimize the return-on-computational-investment (ROCI). Based on both simulated as well as real data, we demonstrate that the proposed optimal HTVS framework can significantly accelerate screening virtually without any degradation in terms of accuracy. Furthermore, it enables an adaptive operational strategy for HTVS, where one can trade accuracy for efficiency.

Determination of nanoparticle size and size distribution is important because these key parameters dictate nanomaterials' properties and applications. Yet, it is only accomplishable using low-throughput electron microscopy. Herein, we incorporate plasmonic-domain-driven feature engineering with machine learning (ML) for accurate and bidirectional prediction of both parameters for complete characterization of nanoparticle ensembles. Using gold nanospheres as our model system, our ML approach achieves the lowest prediction errors of 2.3% and ±1.0 nm for ensemble size and size distribution respectively, which is 3-6 times lower than previously reported ML or Mie approaches. Knowledge elicitation from the plasmonic domain and concomitant translation into featurization allow us to mitigate noise and boost data interpretability. This enables us to overcome challenges arising from size anisotropy and small sample size limitations to achieve highly generalizable ML models. We further showcase inverse prediction capabilities, using size and size distribution as inputs to generate spectra with LSPRs that closely match experimental data. This work illustrates a ML-empowered total nanocharacterization strategy that is rapid (<30 s), versatile, and applicable over a wide size range of 200 nm.

Optimized machine learning approaches for identifying vertical temperature gradient on ballastless track in natural environments

Performance optimization of wireless networks is typically complicated because of high computational complexity and dynamic channel conditions. Considering a specific case, the recent introduction of intelligent reflecting surface (IRS) can reshape the wireless channels by controlling the scattering elements' phase shifts, namely, passive beamforming. However, due to the large size of scattering elements, the IRS's beamforming optimization becomes intractable. In this article, we focus on machine learning (ML) approaches for complex optimization problems in wireless networks. ML approaches can provide flexibility and robustness against uncertain and dynamic systems. However, practical challenges still remain due to slow convergence in offline training or online learning. This motivated us to design a novel optimization-driven ML framework that exploits the efficiency of model-based optimization and the robustness of model-free ML approaches. Splitting the control variables into two parts allows one part to be updated by the outer loop ML approach while the other part is solved by the inner loop optimization. The case study in IRS-assisted wireless networks confirms that the optimization-driven ML framework can improve learning efficiency and the reward performance significantly compared to conventional model-free ML approaches.

Chemical heuristics have been fundamental to the advancement of chemistry and materials science. These heuristics are typically established by scientists using knowledge and creativity to extract patterns from limited datasets. Machine learning offers opportunities to perfect this approach using computers and larger datasets. Here, we discuss the relationships between traditional heuristics and machine learning approaches. We show how traditional rules can be challenged by large-scale statistical assessment and how traditional concepts commonly used as features are feeding the machine learning techniques. We stress the waste involved in relearning chemical rules and the challenges in terms of data size requirements for purely data-driven approaches. Our view is that heuristic and machine learning approaches are at their best when they work together.

Background: Adolescent depression is highly prevalent in low- and middle-income countries (LMICs). Identifying top key risk factors is necessary to inform effective prevention program design. Machine learning (ML) offers a powerful approach to analyze complex multidomain of data to identify the most relevant predictors and estimate risks for mental health problems. This paper applies ML to study risks for adolescent depression to enhance adolescent depression prevention efforts in LMICs. Methods: Six ML approaches (e.g., Explainable Boosting Machine, random forests, and XGBoost) were applied to study the risks of depression. Data were drawn from a digital health intervention study conducted in Kenya (year 2024–2025, n = 269). Multiple domains of childhood and adolescent adversity and stress experiences were used to predict adolescent depression (using PHQ9-A). Findings: We found that ML was a valuable approach in the early identification of adolescents at risk for depression. Among the six ML approaches applied, the random forest approach outperformed other ML approaches, especially when multiple domains of risks were included. We also found that childhood adversity or home adversity alone were not strong predictors for depression. Adding adolescent stress experiences and community school adversity experiences significantly improves the accuracy and predictability of depression. Using the top-15 and top-20 ranking factors, we achieved 74.8% and 75.1% accuracy in depression prediction, which was similar to the accuracy when all 49 adverse/stress factors were included in the predictive model (78.3%). Conclusions: Innovative ML and modern predictive modeling approaches have the potential to transform modern preventive mental health care by better utilizing multidomain data to identify individuals at risk for developing depression early and identify top risk factors (for targeted individuals and/or populations). Findings from ML can inform tailored intervention design to better mitigate risks in order to prevent depression problem development. They can also inform the better utilization of resources to target high-need cases and key determinants, which is particularly useful for LMICs and low-resource settings. This paper illustrates an example of how to move toward this direction. Future research is needed to validate the approach.

Active Learning (AL) is a methodology from machine learning in which the learner interacts with the data source. In this paper, we investigate application of AL techniques to a new domain: regression problems in performance analysis. For computational systems with many factors, each of which can take on many levels, fixed experiment designs can require many experiments, and can explore the problem space inefficiently. We address these problems with a dynamic, adaptive experiment design, using AL in conjunction with Gaussian Process Regression (GPR). The performance analysis process is seeded with a small number of initial experiments, then GPR provides estimates of regression confidence across the full input space. AL is used to suggest follow-up experiments to run, in general, it will suggest experiments in areas where the GRP model indicates low confidence, and through repeated experiments, the process eventually achieves high confidence throughout the input space. We apply this approach to the problem of estimating performance and energy usage of HPGMG-FE, and create good-quality predictive models for the quantities of interest, with low error and reduced cost, using only a modest number of experiments. Our analysis shows that the error reduction achieved from replacing the basic AL algorithm with a cost-aware algorithm can be significant, reaching up to 38% for the same computational cost of experiments.

Surface water pollution become a nuisance for humankind as river water fulfill requirement of a major population and traditional method of water quality assessment and evaluation is inadequate in this era. So using advance method of machine learning in prediction of surface water proves to be helpful to prevent future water accident. As we seen many recent studies of water quality prediction and river water assessment using machine learning approach for better accuracy and less labor and to optimize its overall results. It’s become essential to review the recent studies which used Machine Learning algorithms for prediction, analysis, evaluation and assessment of river water quality and different models used in these studies for different environmental conditions. Machine learning models are superior to handle such complex and non linear data such as water quality parameters with greater accuracy, reliability, cost-effectiveness and efficiency as considered as great tool for surface Water Quality monitoring, prediction, future projects and help lawmakers in policy. In this report we reviewed around 17 research papers which uses machine learning approach from different journal and concise it to covers the structure of study, datasets used, methodology analysis, models performance, environment susceptibility, comparative analysis and assessments of Machine Learning models progress in river water quality research. For better management and control of surface water quality and its treatment, this study will help in understanding and analyzing the studies reviewed in this paper and its future application. We can conclude that research on Water Quality prediction using Machine Learning model are inadequate in the context future vulnerability, observing increasing pollution in recent years we require more research in this field. Finally, this study provides breakthrough in Surface Water Engineering and Management to give a new direction to fore coming studies and fortified it scope also gives a comparative approach for its implementation in new studies.

This study predicts the minimum energy pathways (MEPs) and transition barriers of two-dimensional transition metal dichalcogenides (TMDs) during their 1H to 1T structural phase transitions. The investigation utilizes density functional theory (DFT) calculations and machine learning algorithms to predict the MEPs and transition barriers. Six TMDs, namely NbSe2, ScS2, ScSe2, TiTe2, VS2, and VSe2, are selected for analysis. The DFT calculations provide reference values for comparison with the machine learning predictions. The transition barriers obtained through DFT calculations range from 0.13 eV to 0.72 eV. The machine learning models, including Linear Regression, Random Forest Regression, K-Nearest Neighbor Regression, and Gaussian Process Regression, demonstrate varying degrees of accuracy in predicting the transition barriers. Gaussian Process Regression achieved the highest accuracy among all ML models, with the absolute mean error spanning all systems being 0.0093. It has been shown that the features that were identified in the study are sufficient for the prediction of the MEPs and transition barriers of TMDs using machine learning. Overall, this study contributes to the understanding of TMD phase transitions and provides insights into the integration of machine learning in materials science research.

We consider the problem of A-optimal design of experiment under a Bayesian probabilistic model with both categorical and continuous response variables. The utility function of the local design problem is derived by applying Bayesian experimental design framework. We also develop an efficient optimization algorithm to obtain the local optimal design by combining the particle swarm optimization and the blocked coordinate descent methods. In addition, we discuss two different ways of constructing the global optimal design based on the algorithm for local optimal design. Simulation studies are presented to illustrate the efficiency of our approach.