❌

Normal view

Received β€” 27 May 2026 ⏭

TIGER: Text-Informed Generalized Enzyme-Reaction Retrieval

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Yuhang Zhang Β· Keyan Ding Β· Peilin Chen Β· Han Liu Β· Can Lin Β· Ruixi Chen Β· Shiqi Wang Β· Qi Song
26 May 2026 at 12:00
arXiv:2605.24489v1 Announce Type: new Abstract: Enzyme-reaction retrieval is a fundamental problem in computational biology, underpinning enzyme characterization, reaction mechanism elucidation, and the rational design of metabolic pathways and biocatalysts. As a bidirectional task, it entails both enzyme-to-reaction and reaction-to-enzyme mapping. However, existing approaches suffer from poor generalization across tasks and distributions, with performance highly sensitive to dataset splits and substantial asymmetry between retrieval directions. To address these challenges, we present TIGER, a Text-Informed Generalized Enzyme-Reaction Retrieval framework that leverages protein-to-text generation models to distill textual semantic knowledge from enzyme sequences, providing a generalized representation that bridges enzymes and biochemical reactions. To ensure the quality and reliability of textual semantics, we design a Dynamic Gating Network that adaptively fuses text-derived knowledge with sequence features, enabling more consistent and informative enzyme representations, while a Structure-Shared Feature Projector aligns enzyme and reaction representations within a unified latent space. Extensive experiments demonstrate that, under bidirectional retrieval supervision, TIGER significantly outperforms state-of-the-art baselines across diverse distributions and exhibits strong robustness and transferability across tasks.

Channel-wise Vector Quantization

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Wei Song Β· Tianhang Wang Β· Yitong Chen Β· Tong Zhang Β· Zuxuan Wu Β· Ming Li Β· Jiaqi Wang Β· Kaicheng Yu
26 May 2026 at 12:00
arXiv:2605.26089v1 Announce Type: cross Abstract: We present Channel-wise Vector Quantization (CVQ), a novel image tokenization paradigm that replaces patch-wise tokens with channel-wise tokens. Unlike conventional vector quantization, which assigns a discrete token to each patch feature vector, CVQ quantizes each channel of the feature map. This formulation represents an image as discrete levels of visual details, rather than as a grid of spatial patches. Based on CVQ, we introduce a new visual autoregressive framework with "next-channel prediction". Instead of rendering images patch by patch in raster order, our Channel-wise Autoregressive (CAR) model predicts image channels sequentially, producing progressively enriched visual details. Specifically, it first sketches global structure and then refines fine-grained attributes, akin to a human artist's workflow. Empirically, we show that: (1) CVQ achieves 100% codebook utilization with a 16K+ codebook size without any bells and whistles, and substantially improves reconstruction quality over conventional VQ; and (2) CAR attains a DPG score of 86.7 and a GenEval score of 0.79, demonstrating strong effectiveness for text-to-image generation.

Topology-Driven Transferability Estimation of Medical Foundation Models for Segmentation

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Jiaqi Tang Β· Shaoyang Zhang Β· Xiaoqi Wang Β· Jiaying Zhou Β· Yang Liu Β· Qingchao Chen
26 May 2026 at 12:00
arXiv:2602.23916v2 Announce Type: replace-cross Abstract: The advent of large-scale self-supervised learning (SSL) has produced a vast zoo of medical foundation models. However, selecting optimal medical foundation models for specific segmentation tasks remains a computational bottleneck. Existing Transferability Estimation (TE) metrics, primarily designed for classification, rely on global statistical assumptions and fail to capture the topological complexity essential for dense prediction. We propose a novel Topology-Driven Transferability Estimation framework that evaluates manifold tractability rather than statistical overlap. Our approach introduces three components: (1) Global Representation Topology Divergence (GRTD), utilizing Minimum Spanning Trees to quantify feature-label structural isomorphism; (2) Local Boundary-Aware Topological Consistency (LBTC), which assesses manifold separability specifically at critical anatomical boundaries; and (3) Task-Adaptive Fusion, which dynamically integrates global and local metrics based on the semantic cardinality of the target task. Validated on the large-scale OpenMind benchmark across diverse anatomical targets and SSL foundation models, our approach significantly outperforms state-of-the-art baselines by around 31% relative improvement in the weighted Kendall metric, providing a robust, training-free proxy for efficient model selection without the cost of fine-tuning. The code will be made publicly available upon acceptance.

Integrated multi-omics profiling reveals phenotype- and tissue-specific host-microbiota interactions in paired tumor and peritumoral tissues of advanced gastric cancer patients from Northwest China

βœ‡Omics in Gastric
By: Anqi Wang Β· Qi Wang Β· Ting Zhang Β· Guoqing Qi Β· Wen Ren Β· Wenji Tian Β· Juanjuan Chen
6 May 2026 at 18:00

Front Cell Infect Microbiol. 2026 Apr 20;16:1763765. doi: 10.3389/fcimb.2026.1763765. eCollection 2026.

ABSTRACT

BACKGROUND: Advanced gastric cancer (AGC) exhibits a high incidence in Northwest China, largely attributed to region-specific dietary patterns and environmental exposures. Its pathogenesis involves complex host-microbiota crosstalk, which has not yet been comprehensively elucidated through integrated multi-omics approaches. Herein, we employed trasncriptomic and shotgun metagenomic sequencing on paired tumoral and peritumoal mucosal tissues from 88 AGC patients in Northwest China. Our aim was to systematically characterize host gene expression profiles, the composition and functional potential of the gastric mucosal microbiota, and their intricate interrelationships.

RESULTS: Transcriptomic profiling clearly distinguished tumoral from peritumoral regions (PERMANOVA, R2 = 0.24, P = 0.0001), with 8870 differentially expressed genes (DEGs) identified between the two tissue types. Tumor tissues harbored 8377 up-regulated DEG, which were enriched in extracellular matrix (ECM) organization, cell cycle regulation, signaling transduction, and inflammatory pathways (e.g., PI3K-Akt, IL-17 signaling). In contrast, peritumoral tissues showed 493 up-regulated DEGs primarily associated with metabolic processes. Host gene expression was significantly modulated by Lauren classification in tumoral mucosa (P = 0.025) and by Helicobacter pylori (Hp) infection in peritumoral tissues (P = 0.0424). Hp-infected tissues exhibited 65 up-regulated DEGs linked to transcriptional misregulation in cancer, inflammation, immune activation and mitochondrial pathways. Lauren subtypes displayed distinct transcriptomic signatures: intestinal-type AGC was enriched in metabolic processes, diffuse-type in immune and signal transduction pathways, and mixed-type in Ras/MAPK/ErbB and NF-ΞΊB signaling pathways. Correlation analysis between the 8870 DEGs and seven differentially abundant bacterial species (e.g., Serratia surfactantfaciens, Pseudomonas protegens, Prevotella jejuni, and Streptococcus infantis) revealed 13199 significant correlations. Among these, S. surfactantfaciens and P. protegens exhibited the strongest connectivity with host genes. Functionally, the correlated DEGs were involved in ECM structure, cell cycle progression, immune and inflammatory responses, cellular proliferation and differentiation, and metabolic processes.

CONCLUSIONS: Our findings demonstrated phenotype- and tissue-specific regulation of host gene expression in AGC and revealed extensive host-microbe interactions. This work fills a critical gap in multi-omics research on AGC in the Northwest Chinese population and suggests potential diagnostic and therapeutic targets for AGC.

PMID:42088021 | PMC:PMC13136179 | DOI:10.3389/fcimb.2026.1763765

There are no more articles
❌