时间标签：数据日期 2026-11-05 · 生成时间 2026-11-05 05:30（Asia/Shanghai）｜ 检索窗口 PubMed edat 2026/11/02 – 2026/11/06 ｜ 情报发生日期 2026-11-05
研究问题
We hypothesized that while cross-hybridization among target species is unavoidable when using many species-specific probes in a single microbiome labeling scenario, small differences in hybridization efficiency would be observed when probes cross-hybridize to off-target species with nucleotide mismatches.；For model training and benchmarking, bacterial cells were harvested at mid-log phase and fixed in 4% (w/v) paraformaldehyde (PFA) (EMS Diasum) at RT for 1.5 h.Generation of the controlled bacterial community in gnotobiotic zebrafishTo generate the controlled bacterial community prior to imaging, the relative proportions of individual cells of the 10 species described in Table S1 were estimated using FISH-based image quantification across four fields of view (FOVs) for pure isolates acquired with the Carl Zeiss LSM 980 with the 20×/ 1.4 NA objective.；The images were acquired with the Plan-Apo 63×/Oil DIC 1.4 NA objective in lambda mode using the 32-anode spectral detector to sample multiple 8.9 nm wavelength bands.
方法
The ever-decreasing cost of DNA sequencing has enabled a foundational understanding of the composition of the microbiome among healthy subjects, and changes that occur in community composition due to numerous factors, including diet and lifestyle [5–10].；High-throughput assays including metagenomics, metaproteomics, and metabolomics are providing a path to investigate the interactions among gut microbes and between microbes and their host [11–13].；Nonetheless, these techniques are typically applied to homogenized samples and neglect the spatial heterogeneity that underlies function within the complex metazoan gut ecosystem.The gut community spatial structure, both biogeography and multicellular architecture, dictate host–microbe interactions and community functions [5, 14–16].
核心结论
Visual inspection of images of pure populations after CIPHR-FISH classification demonstrate robust classification of species and strains as evidenced by their isolated appearance along the diagonal in the image matrix plot of species identity vs.；Error bars = standard deviation from n = 4 images.Representative fluorescence images demonstrate CIPHR classification of individually labeled bacterial isolates and a synthetic 10-member bacterial community.；Relative-abundance plots, microbial-load measurements, and a taxon-specific heat map demonstrate regional differences in community composition, with microbial abundance highest anteriorly and decreasing toward the posterior region.To assess the prediction stability of the in vivo taxonomic assignments to variation in the training data, we performed 1000 bootstrap resampling iterations separately for the skin and gut classification models.
局限
Multiple bacterial species were detected across animals in varying proportions; however, S. maltophilia was consistently absent from the skin microbiota.；In this study, we introduce CIPHR-FISH, a probe design and machine learning pipeline that transforms the problem of probe cross-hybridization and spectral crosstalk, traditionally treated as experimental artifacts to be avoided as much as possible, into informative molecular signatures that can be learned from reference images, without the need for sophisticated probe design to enable species-level resolution in microbial 16S FISH.CIPHR-FISH also overcomes other limitations in microbial FISH.；However, the 5S and 23S rRNAs are also present in as many copies.
对本项目的价值
中高价值：属「微生物组及其代谢物干预」方向；微生物组及其代谢物干预栏目可用作本项目文献库与综述素材。 ｜ 价值分 3.0/5（★★★☆☆）
全文来源
PubMed PMC JATS（正文 52,935 字符，题名指纹 0.87，已过 PubMed PMCID 官方关联 + 题名指纹 + 正文长度归属校验）
原文题录
- 中文题名：交叉杂交推断系统发育分辨-荧光原位杂交实现菌株级分类分辨的菌群成像
- 英文标题：Cross hybridization inference for phylogenetic resolution-fluorescence in situ hybridization enables microbiome imaging with strain level taxonomic resolution.
- 期刊与标识：ISME communications · PMID 42979483 · DOI 10.1093/ismeco/ycag296 · PMC PMC13730059