Cancer is a highly heterogeneous disease, characterized by significant variability across multiple dimensions. This diversity has been extensively studied from various perspectives. In this study, we aim to examine cancer heterogeneity through the lens of immune cell composition within the tumor microenvironment. Transcriptome profiles of 11,274 subjects from 33 cancer types were deconvoluted to infer the composition of 22 immune cell types. A deep learning‐based model was adapted to model a binary outcome for cancer type, tumor versus normal. The Shannon index was used to represent immune cell diversity. Cox proportional hazards regression was used to evaluate the prognostic values of immune cells. Tumor and normal tissues showed significantly different immune cell compositions in 183 of 352 comparisons, with macrophage M0 levels consistently elevated in most tumors except lung adenocarcinoma. ResNet models using immune cell data achieved strong performance in distinguishing tumor versus normal samples (average F1 score = 0.64 when combined with clinical features). Four immune features were significantly associated with disease‐specific survival across three cancer types: macrophage M0 and macrophage M2 in bladder urothelial carcinoma, macrophage M0 in kidney renal clear cell carcinoma, and the Treg/CD8⁺ T cell ratio in skin cutaneous melanoma. Although immune cell diversity varied across cancers, it was not broadly predictive of prognosis, highlighting the cancer type‐specific nature of immune remodeling. Our results demonstrated cancer heterogeneity through the analysis of immune cell composition and diversity using deep learning models and prognostic analysis. The significant findings revealed varying patterns across different cancer types, indicating that no unifying standard can be established across all cancers.

The efficient, rapid, and reliable assembly of DNA fragments is essential for advancing metabolic engineering and synthetic biology. With the rapid advancement of DNA synthesis and assembly technologies, the scale of DNA assembly has expanded from single genes to metabolic pathways and even genomes. Large DNA fragment assembly, in particular, has developed into a pivotal tool for microbial cell factory engineering, significantly contributing to the understanding and construction of biological systems. This review summarizes recent advancements in large DNA fragment assembly methods, encompassing in vitro and in vitro methods for multiple‐gene assemblies, as well as genome‐scale technologies, including synthetic genome and neochromosome construction. We systematically compare their key features in terms of assembly principle, capacity, and efficiency. Especially, we highlight their applications in microbial cell factory engineering, including heterologous pathway construction, reprogramming host metabolism, and expanding complex biosynthetic networks. Finally, we discuss the challenges and prospects of applying large DNA fragment assembly to advance cell factories. In summary, this review provides a theoretical and technical framework for engineering high‐performance microbial cell factories, contributing to the advancement of industrial biomanufacturing.

Repetitive DNA, a fundamental architectural element of genomes, is widespread across organisms and comprises about 54% of the human genome. With advances in long‐read sequencing and bioinformatics approaches, highly repetitive sequences can now be characterized in depth. Nevertheless, engineering repetitive DNA is mainly limited by repeat instability. Despite these challenges, advances have been reported in the construction of short tandem repeats and the large‐scale duplication of genes. This review provides a comprehensive overview of current strategies for constructing repetitive DNA, covering both in vitro approaches, including restriction endonuclease ligation, in vitro homologous recombination, and redundancy‐minimizing sequences, as well as in vivo methodologies in Escherichia coli, Saccharomyces cerevisiae, and mammalian cells. We further discuss applications of these methods for developing genetic manipulation tools, constructing models of repetitive DNA, and synthesizing biomaterials. Future progress in engineering repetitive DNA will further advance our capabilities in designing and manipulating complex biological systems and deepen our understanding of them.

Single‐cell multi‐omics sequencing technology provides a powerful tool for studying cellular heterogeneity. However, beyond the challenges of sparsity, heterogeneity, and dimensionality differences, a critical challenge in multi‐omics data integration lies in preserving the true regulatory relationships among molecular features. To address these limitations, we propose single‐cell multi‐omics graph neural networks (scMOG), a framework that leverages heterogeneous graphs to preserve regulatory relationships in single‐cell multi‐omics data. scMOG leverages encoders to extract low‐dimensional embeddings of both cells and features while reconstructing the input data using zero‐inflated negative binomial decoders, effectively handling high sparsity and noise. In addition, scMOG introduces a contrastive learning module and an omics alignment module to preserve differences in expression patterns across distinct omics while extracting consistent information. Experimental results on eight single‐cell multi‐omics datasets demonstrate that scMOG outperforms existing methods, producing embeddings that capture meaningful biological signals. scMOG provides an effective solution for integrating single‐cell multi‐omics data, offering a scalable framework that preserves regulatory signals.

Quantitative analysis of fluorescence microscopy images remains a significant challenge in cytopathological assessments, as subjective interpretation and inter‐observer variability hinder reproducibility and standardization of classifications. In this study, an artificial intelligence‐assisted analytical framework utilizing a multimodal large language model (MLLM) was employed to classify cytotoxic responses in acridine orange/propidium iodide‐stained MCF‐7 cells subjected to doxorubicin treatment for 24 h. A total of 500 fluorescence images, collected from five experimental groups including a control and four dose‐dependent treatment conditions, were examined to identify distinct morphological indicators of viability, necrosis, and apoptosis. Ground truth annotations were established using a standardized 15‐parameter cytopathological evaluation checklist, conducted by domain experts. The MLLM was subsequently tasked with interpreting the same dataset and generating structured cytopathological outputs aligned with expert‐defined criteria. Receiver operating characteristic analysis demonstrated moderate discriminative performance for viable cells (area under the curve [AUC] = 0.68) and strong discriminative performance for necrotic cells (AUC = 0.84), reflecting reliable identification of overt necrotic morphology. However, early and late apoptotic states showed poor separability (AUC = 0.52–0.55) and were inconsistently distinguished, which can be attributed to inadequate spatial–contextual inference within the model. The entire dataset was processed within 2 h, demonstrating a significant enhancement in analytical throughput relative to manual evaluation. These findings indicate that MLLM‐driven cytopathological analysis may serve as an intermediate automation layer between traditional microscopic assessment and fully autonomous deep learning‐based image processing systems, facilitating scalable, standardized, and semi‐autonomous interpretation of fluorescence imaging data.

Zebrafish are a model organism used for the study of vertebrate development, disease, and drug discovery. Two‐day old zebrafish exhibit burst swimming behavior that can be elicited by a light touch to the tail. Motor touch responses are frequently video‐recorded and later analyzed by hand. Despite this being a simple behavioral assay, methods for robustly analyzing these videos in a reproducible and time‐efficient manner rely on manual tracking, which is prone to experimenter bias and error. Here we present LaZeTrack, a machine learning‐based program, which employs Ultralytics’ YOLOv8 nano (YOLOv8n) object detection algorithm to automatically analyze touch response videos using 2‐day‐old zebrafish. This program breaks down video files into their constituent frames which are then passed through a custom‐trained YOLOv8n algorithm to detect the presence of a single zebrafish (mean average precision, mAP50–95 = 0.74). LaZeTrack then refines the tracking data output by the model and tabulates it into an Excel spreadsheet. It then computes and extracts four relevant swim metrics: swim duration (s), swim distance (mm), mean swim velocity (mm/s), and maximum swim velocity (mm/s). LaZeTrack rapidly accelerates the analysis process, while also eliminating errors associated with manual tracking. Furthermore, it allows for high‐throughput analysis of zebrafish touch response videos and can detect subtle differences in motor metrics arising from temperature differences, demonstrating the utility of this tracking algorithm. LaZeTrack is available for download on GitHub (Armstrong‐Lab‐70/LaZeTrack).

Genome–phenome association (GPA) prediction can broaden the understanding of biological mechanisms underlying complex phenotypic traits (e.g., diseases and agronomic traits). Traditional deep matrix factorization (DMF)‐based GPA methods can integrate multiple data types and uncover nonlinear associations but often rely on low‐dimensional matrix representations, making it challenging to capture intricate node interactions under sparse data conditions. Existing approaches also inadequately integrate multi‐omics data and frequently neglect local neighborhood (LN) and graph structural information, particularly when handling incomplete or low‐quality attribute data. Furthermore, most methods fail to effectively address the scarcity of plant phenotype data. We propose GDMF, an end‐to‐end GPA prediction framework that integrates a multisource graph attention network (GAT) with weighted DMF. By constructing a heterogeneous molecular network, GDMF unifies multi‐omics data and utilizes GAT’s attention mechanism to process multisource associations (both cross‐type and intra‐type node interactions), thereby enhancing the capture of LN structures while preserving critical node information. The weighted DMF component dynamically assigns importance weights to global relationship matrices, enabling robust modeling of complex gene–phenotype interactions even with sparse and noisy attribute data. To mitigate the scarcity of plant phenotype data, GDMF incorporates textual semantic similarity of trait ontology terms to enrich phenotypic attribute representations. Extensive experiments on maize and human datasets demonstrate that GDMF significantly outperforms competitive methods, validating its efficacy and superiority in GPA prediction.

In this study, a fractional‐order SE{I}_{u}ITR model was proposed to examine the transmission of tuberculosis (TB) in the community. The proposed model was rigorously examined for well‐posedness. The basic reproduction number was also derived, and the disease‐free equilibrium was obtained. The local and global stabilities of the proposed model were also examined. The proposed model was fitted with the actual TB incidence in Ethiopia, while numerical simulations were carried out in MATLAB to validate the results obtained in this study. Local and global sensitivity analyses were carried out to determine the key parameters that influence the transmission of TB in the community. Phase‐plane portraits and phase‐plane diagrams demonstrate the dynamic process of the state variables. The results indicate that as the fractional‐order increases, the number of infections decreases due to the memory effect of early detection and treatment, while as the order decreases, the number of infections increases. Early detection and treatment have a profound effect on recovery and prevalence.

The skin microecology plays a vital role in maintaining cutaneous health and is intricately linked to host skin phenotypes. However, there remains a lack of precise and quantitative biomarkers for evaluating skin health conditions, making it challenging to identify individuals at potential risk of microecological imbalance. In this study, we collected facial skin microbiomes from 242 female volunteers aged 16–50 years in Shanghai, China. Microbial communities were surveyed using 16S rRNA gene sequencing and integrated with high‐resolution facial images and host skin phenotypes for comprehensive analysis. Our findings reveal that although the community complexity of the facial microbiome is comparable between individuals with “ideal” and “non‐ideal” skin conditions, the composition and structure of key microbes varied significantly between the two groups. Differential abundance analysis further emphasized the role of interactions between the skin microbiome and host phenotypic traits in skin aging and status transitions. To enable a more holistic and quantitative assessment of facial skin status, we developed a novel multimodal skin index (MSI) that fuses multiple modal data from facial images, microbiome profiles, and host skin phenotype features through a deep learning‐based framework. Importantly, MSI identified individuals with an outwardly healthy facial appearance but significant underlying microbial dysbiosis that conventional diagnostic approaches often overlook. This work enables the detection of such hidden risks, offering new avenues for individualized facial skin health assessment, precision dermatology, and microbiome‐informed esthetic interventions.

Single‐cell RNA sequencing (scRNA‐seq) has transformed the study of cellular heterogeneity by making it possible to classify individual cells and their functional states. However, the analysis remains difficult because high dropout rates lead to sparse and noisy expression data. Existing marker gene selection methods are often fragmented, typically relying on a single strategy that overlooks complementary biological signals. Without a unifying framework for denoising and marker identification, clustering can suffer in accuracy, stability, and interpretability. This makes it more difficult to define cell subpopulations and extract meaningful biological insights clearly. We introduce single cell marker gene clustering (scMGC), a unified framework that integrates a denoising autoencoder with multi‐method marker selection for scRNA‐seq clustering. The autoencoder reduces dropout noise, whereas a unified marker gene scoring system balances the contributions of multiple methods to select reliable markers. These markers are then applied in graph‐based clustering to uncover accurate and interpretable cell subpopulations. scMGC was benchmarked against seven state‐of‐the‐art methods across seven scRNA‐seq datasets. On average, it outperformed competing approaches by 31.3% in adjusted rand index and 28.2% in normalized mutual information, showing consistent improvements in clustering accuracy. Enrichment and disease association analyses further validated that the discovered clusters are both biologically meaningful and clinically relevant. The codes and datasets used are available on the GitHub website (srniloy/scMGC).