A Computational Approach to Statistical Learning

The placebo and nocebo effects highlight the importance of expectations in modulating pain perception, but in everyday life we don’t need an external source of information to form expectations about pain. The brain can learn to predict pain in a more fundamental way, simply by experiencing fluctuating, non-random streams of noxious inputs, and extracting their temporal regularities. This process is called statistical learning. Here we address a key open question: does statistical learning modulate pain perception? We asked 27 participants to both rate and predict pain intensity levels in sequences of fluctuating heat pain. Using a computational approach, we show that probabilistic expectations and confidence were used to weight pain perception and prediction. As such, this study goes beyond well-established conditioning paradigms associating non-pain cues with pain outcomes, and shows that statistical learning itself shapes pain experience. This finding opens a new path of research into the brain mechanisms of pain regulation, with relevance to chronic pain where it may be dysfunctional.

Biomass-derived carbon materials have gained attention for their potential in energy storage applications, mainly due to their sustainability and cost-effectiveness. Among various sources, coconut shells offer high carbon content and are widely available. However, optimizing the activation process remains a key challenge in enhancing electrochemical performance. This study investigates the effects of particle size, activation temperature, activation time, and CSAC to potassium bicarbonate (KHCO₃) ratio on the specific capacitance of the electrode material. To better understand and predict the relationships between processing conditions and performance, a combination of statistical analysis and machine learning methods were used. These included correlation analysis, multivariate regression, Random Forest regression, and Response Surface Methodology (RSM). Results showed that activation temperature and time had the most significant influence, contributing to about 80% of the variance in capacitance. The best outcome of Run 52, was achieved using a 1:2 ratio (CSAC: KHCO3), with optimal activation conditions of 800 °C and 120 minutes, yielding a maximum actual specific capacitance of 298 F g⁻¹. The Run 52-CSAC attained 91.2 % capacitance retention at 1 A/g current density after 10,000 cycles. The Random Forest model performed well (R² = 0.913, RMSE = 8.21 F g⁻¹), closely aligned with regression results (R² = 0.904, p < 0.001). RSM analysis also highlighted a nonlinear interaction between temperature and activation ratio. This integrated experimental and computational approach provides a reliable framework for optimizing activation processes in biomass-derived carbons. However, findings are limited to a single biomass type and activation method, suggesting the need for further studies on different materials and conditions to validate broader applicability.

Tenofovir and entecavir are currently designated as the preferred oral antiviral drugs for chronic hepatitis B. However, only less than 40% of patients can achieve HBeAg seroconversion. We aim at investigating the role of intestinal microbiome in HBeAg seroconversion induced by oral antiviral therapy and describe multi-omics characteristics of HBeAg seroconversion associated intestinal flora. In this study, we prospectively collected fecal samples at baseline from the patients with HBeAg positive chronic hepatitis B who would have oral antiviral therapy. 16S rDNA sequencing and metabolomics were performed. We identified HBeAg seroconversion-related microbial signature and constructed prediction model for HBeAg seroconversion. Thirty-seven of these subjects achieved HBeAg seroconversion within 156 weeks after the initiation of oral antiviral therapy, while 41 subjects remained HBeAg positive even after over 156 weeks of therapy. A computational statistical and machine learning approach allowed us to identify a microbial signature for HBeAg seroconversion. Using random forest method, we further constructed a classifier based on the microbial signature, with area under curve being 0.749 for the test set. Patients who achieved HBeAg seroconversion tended to have lower abundance of certain fecal metabolites such as essential amino acids, and several dipeptides. By analyzing the fecal microbiota from the patients with and without HBeAg seroconversion, we showed intestinal microbiome play a critical role in HBeAg seroconversion induced by oral antiviral therapy. We also identified intestinal microbial signature that is associated with HBeAg seroconversion after oral antiviral therapy.

Toward a Revised Nosology for Attention-Deficit/Hyperactivity Disorder Heterogeneity

Introduction: Supervised statistical learning for cell-level segmentation and morphometry in optical microscopy is limited less by algorithmic capacity than by the scarcity of reliable, expert-validated ground truth. In comparative neuroscience and quantitative histology, where classical stains such as Nissl's method remain the primary means to study cellular morphology, this bottleneck is acute: manual annotation is expensive, subject to individual bias, and rarely performed at the scale or consistency that computational approaches demand. No existing platform integrates a stain-specific bioimage segmentation protocol, a structured multi-annotator workflow, and consensus-based quality control into a single pipeline from image ingestion to machine-readable training data. Methods: We present Anatolution, an open-source, web-based platform designed to address the gap of quality annotations at https://anatolution.herokuapp.com/public-tool/. Anatolution organizes microscopy images, including 2D arrays or 3D volumes, into project workspaces where multiple annotators independently label cellular structures against a shared computer vision catalogue. This design enables systematic inter-rater and intra-rater reliability assessment, with consensus derived from agreement across annotators rather than from any single expert's judgment. The platform enables the export of aggregated labels or annotation datasets for downstream statistical learning methods. We describe the system's architecture, its Nissl-specific segmentation pipeline, the consensus annotation workflow, and validation of inter-rater reliability. Conclusion: Across 20+ histological annotation containers annotated by up to 15 independent raters, consensus boundary agreement increased monotonically with annotator count, reaching a median Dice of 0.79 against the full-rater reference at seven annotators, with top-tier containers achieving leave-one-out ceiling values of 0.621 to 0.769 for cell-body segmentation. The segmentation pipeline provided effective spatial anchoring, with 88% of consensus-annotated polygons containing at least one algorithmically detected seed. Anatolution provides open-source infrastructure for producing consensus-validated training data from classical histological preparations, addressing the primary bottleneck limiting supervised learning for cell-level morphometry.

Most proteins undergo some form of modification after translation, and phosphorylation is one of the most relevant and ubiquitous post-translational modifications. The succession of protein phosphorylation and dephosphorylation catalyzed by protein kinase and phosphatase, respectively, constitutes a key mechanism of molecular information flow in cellular systems. The protein interactions of kinases, phosphatases, and their regulatory subunits and substrates are the main part of phosphorylation networks. To elucidate the landscape of phosphorylation events has been a central goal pursued by both experimental and computational approaches. Substrate specificity (e.g., sequence, structure) or the phosphoproteome has been utilized in an array of different statistical learning methods to infer phosphorylation networks. In this chapter, different computational phosphorylation network inference-related methods and resources are summarized and discussed.

Bayesian inference is an effective approach for solving statistical learning problems, especially with uncertainty and incompleteness. However, Bayesian inference is a computing-intensive task whose efficiency is physically limited by the bottlenecks of conventional computing platforms. In this paper, a spintronics-based stochastic computing (SC) approach is proposed for efficient Bayesian inference. The inherent stochastic switching behaviors of spintronic devices are exploited to build a stochastic bitstream generator (SBG) for SC with hybrid CMOS/magnetic tunnel junction (MTJ) circuits design. Aiming to improve the inference efficiency, an SBG sharing strategy is leveraged to reduce the required SBG array scale by integrating a switch network between SBG array and SC logic. A device-to-architecture level framework is proposed to evaluate the performance of spintronics-based Bayesian inference system (SPINBIS). Experimental results on data fusion applications have shown that SPINBIS could improve the energy efficiency about 12× than MTJ-based approach with 45% design area overhead and about 26× than FPGA-based approach.

To provide a comprehensive analysis of small molecule genotoxic potential we have developed and validated an automated, high-content, high throughput, image-based in vitro Micronucleus (IVM) assay. This assay simultaneously assesses micronuclei and multiple additional cellular markers associated with genotoxicity. Acoustic dosing (≤ 2 mg) of compound is followed by a 24-h treatment and a 24-h recovery period. Confocal images are captured [Cell Voyager CV7000 (Yokogawa, Japan)] and analysed using Columbus software (PerkinElmer). As standard the assay detects micronuclei (MN), cytotoxicity and cell-cycle profiles from Hoechst phenotypes. Mode of action information is primarily determined by kinetochore labelling in MN (aneugencity) and γH2AX foci analysis (a marker of DNA damage). Applying computational approaches and implementing machine learning models alongside Bayesian classifiers allows the identification of, with 95% accuracy, the aneugenic, clastogenic and negative compounds within the data set (Matthews correlation coefficient: 0.9), reducing analysis time by 80% whilst concurrently minimising human bias. Combining high throughput screening, multiparametric image analysis and machine learning approaches has provided the opportunity to revolutionise early Genetic Toxicology assessment within AstraZeneca. By multiplexing assay endpoints and minimising data generation and analysis time this assay enables complex genotoxicity safety assessments to be made sooner aiding the development of safer drug candidates.