The detection of circulating tumor cells (CTCs) through liquid biopsy offers a non-invasive approach for accurately monitoring cancer dissemination and evaluating therapeutic efficiency. However, their rarity and heterogeneity limit conventional tumor antigen labelling-based methods in identifying and tracing CTCs. Here, we developed a novel metric, termed chromatin unwinding state (CUS), which leverages activated transcriptional regions related to cell-identity processes from single-cell transcriptomic data while overcoming technical variances. Using CUS features, we trained attention-based neural network models, panCTC, to in situ identify and lineage trace rare single CTCs directly from 5 mL of peripheral blood mononuclear cells scRNA-seq without enrichment. We benchmarked panCTC on various in silico-simulated, public, and in-house sequenced data, demonstrating its robustness across sample types and platforms. PanCTC could provide real-time scRNA-seq profiles of fresh CTCs, supporting early cancer detection and targeted anti-metastatic therapy.

Spatial multi-omics defines a DLL4–NOTCH3–driven capillary–mCAF axis orchestrating fibrotic remodeling and immune exclusion in hepatocellular carcinoma tumor cores, conferring immune checkpoint blockade resistance. Targeting stromal NOTCH3 disrupts this pro-fibrotic niche, enhances T cell infiltration, and synergizes with anti-PD-1 therapy, highlighting a translatable strategy to overcome immunotherapy resistance.

Gut microbiome alterations are linked to hepatocellular carcinoma (HCC), yet taxon-based biomarkers often lack reproducibility. Using shotgun metagenomics from 120 resectable HCC (MT) and 76 benign hepatic tumor controls (BT), we reconstructed a stable interaction-based Two Competing Guilds model (HCC-TCG) comprising 142 genomes organized into two antagonistic guilds. The ecological guild balance distinguished HCC from benign and other hepatic tumor types, generalized across external cohorts, and predicted post-resection recurrence and adjuvant therapy response. Validation in advanced HCC immunotherapy cohorts further supported its transferability, establishing a genome-resolved ecological signature for tumor stratification and noninvasive risk prediction.

The study presents a large-scale single-cell transcriptomic atlas profiling over 612,010 peripheral immune cells from 97 individuals to decode the heterogeneity of influenza infection. These findings indicate a fundamental immune dichotomy determining clinical trajectories: a protective, monocyte-centric antiviral state in mild disease versus a pathological, neutrophil- and myeloid-derived suppressor cell (MDSC)-driven hyperinflammatory state in severe infection. Mechanistically, severe disease is orchestrated by the S100A8/9/12–TLR4/RAGE axis and MDSC expansion, driving a “double hit” of cytokine storm and immune paralysis. This dysregulated environment precipitates metabolic collapse in effector T cells and the pathogenic conversion of regulatory T cells. Collectively, these results define the cellular checkpoints governing influenza outcomes and highlight host-directed therapeutic targets.

Immune checkpoint inhibitors (ICIs) have shown promising antitumor efficacy in certain types of solid tumors. However, the efficacy of ICIs remains unsatisfactory owing to the dysregulation of signaling pathways in local tumor tissues. Here, we reveal that diacylglycerol kinase α (DGKα)-derived phosphatidic acid (PA) directly binds to nuclear factor-κB (NF-κB) and enhances the transcriptional activity of NF-κB to increase the expression of programmed cell death 1-ligand 1 (PD-L1) and facilitate the immune evasion of tumor cells and orchestrate immune microenvironment. Inhibition of DGKα activity decreases the intratumoral PD-L1 level and induces cytotoxic T lymphocytes (CTLs) infiltration and resultantly enhances the antitumor efficacy of ICIs. Plasma PA can function as a biomarker to evaluate the efficacy of ICIs in gastrointestinal cancers. Overall, our results identify the DGKα/PA axis as a metabolic driver of immune evasion and CTLs exclusion, representing a promising target to enhance ICIs' efficacy in gastrointestinal cancer treatments.

The graphical abstract illustrates the molecular mechanism by which chondroitin sulfate (DCS) alleviates glucocorticoid-induced myopathy through the gut–muscle axis. DCS selectively enriches L. johnsonii Z-RW, enhancing bshA-encoded BSH activity to promote bile acid deconjugation and reduce the sugar–bile acid ratio, thereby re-establishing intestinal metabolic homeostasis. The resulting secondary bile acids cross the intestinal barrier and reach muscle tissue, where they activate NMRK2-mediated NAD⁺ biosynthesis and PAX7 signaling, upregulating myogenic proteins such as KRT18, MYHC, and RGN. These coordinated molecular events restore energy metabolism and muscle function.

Traditional 16S rRNA gene and Internal Transcribed Spacer region amplicon sequencing provides only relative abundance, often leading to biased ecological interpretations. To overcome this limitation, we developed Accu16S/AccuITS, an absolute quantification method for bacterial and fungal amplicons based on synthetic internal spike-in DNA with known copy numbers. By adding internal standards prior to Polymerase Chain Reaction and sequencing, absolute microbial abundances can be calculated using standard curve regression. Accu16S/AccuITS exhibits sensitivity and consistency comparable to quantitative Polymerase Chain Reaction and is applicable to diverse sample types. A single sequencing run simultaneously yields relative abundance, total absolute abundance, and taxon-specific absolute abundance. Case studies across diverse ecosystems demonstrate that absolute quantification provides ecologically and functionally meaningful insights beyond those obtained from relative abundance analyses.

The figure shows side-by-side quality comparison of the data before and after preprocessing using fastp 1.0. As can be seen from this figure, the data quality has been significantly improved after fastp processing. The Q20 ratio and Q30 ratio increased, and the GC base ratio returned to balance.

EasyMetagenome is a user-friendly shotgun metagenomics pipeline designed for comprehensive microbiome analysis, supporting quality control, host removal, read-based, assembly-based, binning, genome and pan-genome analysis. It offers customizable settings, data visualizations, and parameter explanations. The pipeline is freely available at https://github.com/YongxinLiu/EasyMetagenome.

The concept of “gut–X axis”: the intestine and intestinal microbiota are proven to be able to modulate the pathophysiologic progressions of the extraintestinal organs' diseases. The bioactive chemicals and/or intestinal immune cells can translocate into the circulatory system and other organs and influence the immune reactions, metabolic status, cells physiology, and so forth of extraintestinal organs, finally regulating these organs' homeostasis. Meanwhile, other organs may reversely impact the intestine, namely such regulatory axis is bidirectional.

Fastp is a widely adopted tool for FASTQ data preprocessing and quality control. It is ultrafast and versatile and can perform adapter removal, global or quality trimming, read filtering, unique molecular identifier processing, base correction, and many other actions within a single pass of data scanning. Fastp has been reconstructed and upgraded with some new features. Compared to fastp 0.20.0, the new fastp 0.23.2 is even 80% faster.

Representative visualization results of ImageGP. ImageGP supports 16 types of images and four types of online analysis with up to 26 parameters for customization. ImageGP also contains specialized plots like volcano plot, functional enrichment plot for most omics-data analysis, and other 4 specialized functions for microbiome analysis. Since 2017, ImageGP has been running for nearly 5 years and serving 336,951 visits from all over the world. Together, ImageGP (http://www.ehbio.com/ImageGP/) is an effective and efficient tool for experimental researchers to comprehensively visualize and interpret data generated from wet-lab and dry-lab.

The ggtree object is designed to store phylogenetic tree and associated data, and the object itself is a graphic object that can be rendered as an image file. This work will increase the reproducibility and reusability of phylogenetic data, as well as facilitate integrative and comparative studies.

A new release of PhyloSuite, capable of conducting tree-based analyses. Detailed guidelines for each step of phylogenetic and tree-based analyses, following the “What? Why? and How?” structure. This protocol will help beginners learn how to conduct multilocus phylogenetic analyses and help experienced scientists improve their efficiency.