Tosti et al. report a method for genome-wide, epigenomic profiling that enriches DNA containing unmodified CpG sites for analysis. The enzymatic approach does not damage DNA, nor does it rely on base conversion, and it allows profiling of cell-free DNA with as little as 1 ng of input material.

Reference-guided methods are common in human genomics but not used to analyze microbial community data. Luan et al. describe an approach for reference-guided assembly of microbiome samples and identifying which genomes among the many in public databases can be effectively used to guide the analysis.

Mülling et al. present a spectral flow cytometry-based workflow to profile the metabolism of rare, antigen-specific CD8+ T cells. This approach enables insights into how these immune cells adapt their energy use during infection and cancer, offering tools to better understand disease and guide therapies.

In this review, Ru et al. present actionable strategies for in silico DTI/DTA prediction, covering dataset curation, true-negative construction, and rigorous cold-start evaluation. By integrating multimodal data, heterogeneous networks, LLMs, and AlphaFold, this work outlines a pathway toward more reliable, interpretable, and translational drug discovery pipelines.

Wang et al. develop a statistical physics approach based on the random-field O(n) model (RFOnM) to integrate multiple data types for disease-module detection. Through applications to a variety of complex diseases, they demonstrate that the integrated RFOnM approach presents advantages over single data-type analyses.

Becker et al. develop the Python-based software framework PyRID, which is optimized for the efficient simulation of complex biological molecular systems. PyRID supports diverse features, including multiple molecular reactions, membrane-associated processes, polydispersity, and dense molecular environments. PyRID is designed to facilitate accessibility, extensibility, and seamless integration into diverse research workflows.

Zhang et al. develop a system that reads activity from individual neurons in real time and uses holographic light patterns to control specific neurons in mice, enabling automated, all-optical experiments to study brain circuits during behavior.

Garcia et al. present RESPAN, a pipeline that enables automated dendritic spine analysis through seamless integration of deep learning image restoration and segmentation capabilities with comprehensive quantitation. This streamlined approach ensures accurate and unbiased measurements, even under challenging imaging conditions. It also outperforms manual analysis and existing tools while democratizing access to a combination of advanced capabilities within a single application.

Tai et al. introduce CellSexID, a machine-learning framework for in silico tracking of cell origin in sex-mismatched single-cell RNA-seq data. By leveraging sex prediction as a surrogate, it bypasses experimental labeling and enables accurate origin tracking across chimeric models, organ transplantation, and sample demultiplexing.

Imoto presents an upgraded RECODE platform that simultaneously removes technical and batch noise while preserving full-dimensional single-cell data. Applicable to transcriptomic, epigenomic, and spatial datasets, RECODE improves accuracy, speed, and compatibility with existing workflows, enabling reliable detection of rare cell types and subtle biological variations.

Imani Farahani et al. present a customizable and scalable method to grow neural tube-like tissues from human pluripotent stem cells using a 3D bioprinter. This approach reveals clinical features seen in patients with tuberous sclerosis, offering a high-throughput screening platform to identify disease-causing genes and accelerate drug discovery.

Wang et al. introduce SpliPath, a method to discover gene splicing defects associating with human disease using rare variant analyses. SpliPath combines independent genetic and transcriptomic datasets with sequence-to-function AI models, providing a tool to enhance discovery in rare variant disorders such as ALS.

Ameen et al. introduce Kontextual, a method to evaluate cell interactions within a specified biological context. By doing so, Kontextual produces spatial statistics that account for tissue structures and the choice of imaging region. Kontextual reveals spatial relationships, which are biologically meaningful and can improve the prediction of patient outcomes.

Zhang et al. present Governome, a framework that empowers individuals with comprehensive control over their genomic data. Governome ensures that data remain secure, private, and verifiable by leveraging modern cryptographic technologies. This system protects data integrity and allows for dynamic permission management, providing owners with full transparency over data usage.

Eling and Dorier et al. present standardized workflows for multiplexed immunofluorescence and imaging mass cytometry, enabling large-scale profiling of thousands of cancer tissues in the pan-European IMMUcan consortium. The authors demonstrate high cross-technology consistency and release open-source tools and datasets to support community-wide immune landscape studies.

Guo et al. evaluate current RNA modification calling models used in nanopore sequencing on native and IVT RNA from E. coli. To address the false positives arising from these, they develop nanoSundial, a comparative modification detection tool for identifying prokaryotic RNA modifications.

Wang et al. introduce UniFORM, a universal normalization pipeline for multiplex tissue imaging. UniFORM applies automatic rigid landmark registration to align intensity distributions at both the feature and pixel levels while preserving biological signals. Benchmarking across three platforms demonstrates superior correction of batch effects and improved biological fidelity over existing methods.

Luo et al. introduce APPLE-MS, a method integrating affinity purification and PafA-mediated proximity labeling to overcome limitations in detecting weak, transient, and membrane-associated protein interactions. The approach reveals dynamic viral-host interactomes and endogenous protein networks, offering broad applications in interactome research and therapeutic target discovery.

Liu et al. introduce SuperGLUE, a unified framework for integrating multi-modal single-cell data using probabilistic deep learning and graph-based explainability. It outperforms existing methods in preserving global and local structures, inferring regulatory networks, and enabling interpretable feature relationships through statistical testing, supporting biological discovery across diverse omics modalities.

Lin et al. present a self-driving light-sheet microscope that integrates whole-organ surveys with automated, high-resolution targeting of either user-specified or autonomously identified features. By bridging macroscale mapping and subcellular detail without manual intervention, it offers scalable imaging for developmental studies, neural circuitry mapping, and cancer research.

Zhou et al. present QRscore, a robust non-parametric framework for statistical hypothesis testing that detects shifts in both the mean and variance of gene expression. QRscore maintains high power and strict FDR control, even under noise and zero inflation. It reveals age-related variance changes in both bulk RNA-seq and pseudo-bulked single-cell RNA-seq data.

Ibáñez Alcalá et al. developed a highly adaptable implant, usable for chronic recordings in rats with high precision of brain targeting. Fundamental ideas that are implemented include modularity, precise vertical adjustment, adaptability to different recording systems, accessibility prioritization, and a comprehensive pipeline from building to implantation that could guide future implant development.

Canela et al. introduce a data-driven pipeline to dissect second-order conditioning (SOC) behavioral responses in mice. Their work reveals that SOC elicits unique behavioral signatures depending on the stimulus, highlights age-related deficits in SOC, and demonstrates the power of advanced unsupervised behavioral analysis for uncovering subtle behavioral responses.

Barkai et al. introduce BAREfoot, a machine learning algorithm that combines pose estimation with light-based elevation metrics to automatically score behaviors in rodents. The algorithm enables accurate, high-throughput, and objective detection of pain bouts, offering a powerful tool for behavioral analysis and drug discovery.