把基因表達放回組織原本的座標——15 個步驟帶你掌握空間轉錄組分析，從 Visium 到 Xenium、從 spot 到 subcellular，全程互動模擬與雙語程式碼。
Put gene expression back into its tissue context — 15 chapters covering spatial transcriptomics from Visium to Xenium, from spot to subcellular resolution, with interactive simulations and bilingual code.
什麼是空間轉錄組學？為什麼需要把基因表達跟組織座標一起看？scRNA-seq 與 ST 的核心差異。
What is spatial transcriptomics? Why look at expression with tissue coordinates? Key differences from scRNA-seq.
Spot-based (Visium / Visium HD / Slide-seq / Stereo-seq) vs Image-based (Xenium / MERFISH / CosMx / STARmap)。
Spot-based (Visium / Visium HD / Slide-seq / Stereo-seq) vs Image-based (Xenium / MERFISH / CosMx / STARmap).
spot/cell 層級指標、組織覆蓋、SpotSweeper 局部離群偵測——空間 QC 跟單細胞不一樣。
Spot/cell metrics, tissue coverage, SpotSweeper local outlier detection — spatial QC differs from scRNA-seq.
LogNormalize、SCTransform、scran size factors——spot 含多細胞時應如何選擇？
LogNormalize, SCTransform, scran size factors — how to choose when each spot contains multiple cells?
空間 HVG 篩選、PCA、UMAP，並認識 BANKSY 把空間鄰域納入嵌入。
Spatial HVG selection, PCA, UMAP — and BANKSY embedding that incorporates spatial neighborhoods.
BayesSpace、SpaGCN、STAGATE、GraphST、BANKSY——把組織分成有生物意義的空間區塊。
BayesSpace, SpaGCN, STAGATE, GraphST, BANKSY — partition tissue into biologically meaningful regions.
Moran's I、SpatialDE、SPARK-X、nnSVG——找出表達跟空間位置真正相關的基因。
Moran's I, SpatialDE, SPARK-X, nnSVG — find genes whose expression truly depends on location.
cell2location、RCTD、CARD、SPOTlight、Tangram——估算每個 spot 內的細胞類型組成。
cell2location, RCTD, CARD, SPOTlight, Tangram — estimate cell type composition per spot.
Tangram、CytoSPACE、Seurat label transfer——將 scRNA-seq 細胞類型映射回空間。
Tangram, CytoSPACE, Seurat label transfer — map scRNA-seq cell types back into space.
CellChat v2 spatial、COMMOT、NicheNet——把距離納入 ligand-receptor 推論。
CellChat v2 spatial, COMMOT, NicheNet — incorporate distance into ligand-receptor inference.
CellCharter、NicheCompass、scNiche、Squidpy neighborhood enrichment——量化組織微環境。
CellCharter, NicheCompass, scNiche, Squidpy neighborhood enrichment — quantify tissue microenvironments.
ONTraC niche trajectory、SpaceFlow、stLearn——把 pseudotime 推進空間維度。
ONTraC niche trajectory, SpaceFlow, stLearn — extend pseudotime into the spatial dimension.
Harmony、PASTE、STAligner、STAIR——多切片整合與 3D 重建。
Harmony, PASTE, STAligner, STAIR — multi-slice integration and 3D reconstruction.
Cellpose、Baysor、proseg——image-based 平台 (Xenium / MERFISH / CosMx) 的核心前處理。
Cellpose, Baysor, proseg — critical preprocessing for image-based (Xenium/MERFISH/CosMx) platforms.
2/8/16 µm bin 選擇、HEST 資料庫、視覺 foundation model 把組織形態與表達一起學。
2/8/16 µm bin selection, HEST dataset, vision foundation models that learn morphology + expression jointly.
空間轉錄體學各章對應的工具原始論文、平台規格書、最佳實踐綜述、公開資料集 / Atlas 連結與延伸討論。
Original tool papers, platform specifications, best-practice reviews, open dataset / atlas links, and extended notes for all spatial transcriptomics chapters.
涵蓋各章主題的單選、多選、是非題，內建錯題複習、隨機練習與進度紀錄。
Single-choice, multi-select, and true/false questions across all chapters — with wrong-answer review, random practice, and progress tracking.