Dragonfly 3D World News

关于 Dragonfly 3D World 的新闻与学术动态,每日中英双语速览 Daily bilingual digest of Dragonfly 3D World news and academic updates

📰 2026 年 10 月 24 日日报📰 Digest — October 24, 2026

共收录 2 条:学术论文 2 条 · 官方动态无新增 2 items: 2 research papers · no new official updates

俄罗斯鲟胸鳍棘层状骨的纳米尺度表征:三维 FIB-SEM 研究Nanoscale Characterization of Microscopically Layered Bone in the Russian Sturgeon Pectoral Fin Spine: A 3D FIB-SEM Study

Animals(MDPI)Animals (MDPI)2026-10-21学术论文Research paperDOI: 10.3390/ani16203318

希伯来大学兽医学院团队用 Zeiss Crossbeam 540 FIB-SEM(像素约 10 nm)对俄罗斯鲟(Huso gueldenstaedtii)胸鳍棘的骨单位骨和外周骨做三维成像。图像堆栈导入 Dragonfly 3D World 2022.2 做漂移校正;随后在 Dragonfly 的 Segmentation Wizard 中训练 2D U-Net 模型分割胶原纤维与孔隙,用 Segmentation Comparator 评估(Dice 0.56–0.66),再按特征向量(Eigenvector)各向异性分析绘制钙化胶原纤维束取向图,并对孔隙做 6-连通组件、厚度网格和骨架分析。结果显示:皮质骨单位中纤维取向呈顺/逆时针“扇形”交替,符合板层骨;外周骨虽低倍下看似分层,纳米尺度却以沿棘长轴的单向纤维为主,属平行纤维骨,并伴有成群的孔隙结构。A Hebrew University veterinary team imaged osteonal and peripheral bone of the Russian sturgeon (Huso gueldenstaedtii) pectoral fin spine with a Zeiss Crossbeam 540 FIB-SEM (~10 nm pixels). Stacks were drift-corrected in Dragonfly 3D World 2022.2, then segmented with a 2D U-Net trained in Dragonfly's Segmentation Wizard (evaluated with the Segmentation Comparator, Dice 0.56–0.66); Eigenvector-based anisotropy analysis mapped calcified collagen fibril bundle orientation, and voids were analysed with 6-connected components, thickness meshes and skeleton graphs. Dermal osteons show clockwise–anticlockwise fanning consistent with lamellar bone, while the peripheral bone, layered-looking at low magnification, is mainly unidirectional parallel-fibered bone with grouped voids.

鱼类骨骼Fish boneFIB-SEMFIB-SEM深度学习分割Deep-learning segmentation

用于冷冻光电关联显微(cryo-CLEM)的荧光生物传感器Fluorescent Biosensors for Cryogenic Correlative Light and Electron Microscopy

bioRxiv(预印本)bioRxiv (preprint)2026-10-19学术论文Research paperDOI: 10.64898/2026.10.15.759416

斯坦福大学/SLAC 等团队把钙、pH 和渗透压荧光生物传感器引入冷冻光电关联显微流程,先在低温下标定传感器,再分别用于酿酒酵母、人 U2OS 细胞和细菌 Halothiobacillus neapolitanus。冷冻电子断层(cryo-ET)重建的断层图在 Dragonfly(v2024.1)中用自训练神经网络分割核糖体、细胞壁、羧酶体外壳与内腔,并在 Dragonfly 中手动清理后用 ChimeraX 可视化,把局部生理状态与原位三维结构联系起来。Stanford/SLAC-led researchers bring calcium, pH and osmotic-stress fluorescent biosensors into a cryo-CLEM workflow, calibrating them under cryogenic conditions and applying them in Saccharomyces cerevisiae, human U2OS cells and the bacterium Halothiobacillus neapolitanus. Cryo-ET tomograms were segmented in Dragonfly (v2024.1) with a custom-trained neural network (ribosomes, cell wall, carboxysome shells and lumen), manually cleaned in Dragonfly and visualised in ChimeraX, linking local physiology to in situ 3D structure.

冷冻电子断层Cryo-ET光电关联显微CLEM神经网络分割Neural-network segmentation