# 2.2.5 研究 | 2020-原汁原味读--单细胞肿瘤免疫图谱 > 探索一种新的文献阅读方法，可以更快去理解到作者的主体思路 下面我会从这几块介绍：SOURCE（文章来源）、WHY（作者为什么做这个项目）、HOW（作者怎么做的项目）、GET WHAT（作者得到了什么主要结论） > > 其中不会有特别复杂的词语和语法，而且我会尽可能把文章逻辑层级写清楚 ## SOURCE Title: A Single-Cell Tumor Immune Atlas for Precision Oncology Date: 2020-10-26 Team: Barcelona Institute of Science and Technology (BIST), Barcelona, Spain Paper Link: Data Link (Restricted Access): Code Link: ## WHY？ * Immune microenvironments vary profoundly between patients and biomarkers for prognosis and treatment response lack precision * To pinpoint predictive cellular states of tumor immune cells and their spatial localization ## HOW？ * Analyzing **>500,000** cells from **217** patients and **13** cancer types * Data projection: Seurat's anchor-transferring method * Using **SPOTlight** to combine single-cell and spatial transcriptomics data and identifying striking spatial immune cell patterns in tumor sections * **ShinyApp** (in progress) to project external data and to apply the immune classifier ## GET WHAT? ### **GET 1: Generating a tumor immune cell atlas** * Collected scRNA-seq datasets from **13 different cancer types**, 217 patients and 526,261 cells ![image-20201030104337329](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-024337.png) * Immune cells clustered by cell identity rather than patient origin: integrated **317,111** immune cells using **canonical correlation analysis** => **25** clusters ![image-20201030104248148](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-024248.png) ### **GET2: Tumor subtype classifier** * **For Current:** to establish a pan-cancer immune classification system * used **immune cell type and state frequencies** of the reference atlas as input for similarity assessment across the 13 cancer types ![image-20201030105045062](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-025045.png) * A **hierarchical k-means clustering** using immune cell proportions as features defined six clusters with largely different compositions (almost all cancer types were presented in each cluster) ![image-20201030105413983](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-025414.png) * **For future**: to facilitate the classification of immune profiles * trained an **RF(random forest) classifier** with the 25 immune cell population achieving a highly accurate classification * using the classifier, the pan-cancer immune classification system could be extended to **additional cancer types** ### **GET 3: A resource for immune cell annotation** > To demonstrate the potential value of the atlas * The applicability of the atlas as reference **across different cancer types** * **First:** Project cells onto atlas using a reference-based projection (**Fig. A**) * **Next**: Typical clustering matching (**Fig. B**) ![image-20201030111226885](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-031227.png) * **Third**: Check correlation (**Fig. C**) ![image-20201030112148461](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-032149.png) * The applicability of the atlas as reference **across species** * two liver metastases derived from **mouse CRC organoids** * main subtypes and specific subpopulations could also be assigned **using the human reference** ![image-20201030112314187](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-032314.png) ### **GET 4: Spatial localization of immune cells in tumor sections** > **Spatial** distribution of immune cells is important for **ICI (immune checkpoint inhibitors) response** * Single-cell reference atlas immune profiles + Spatial transcriptome data * **SPOTlight** : non-negative matrix factorization (NMF) based spatial deconvolution framework * Analysis of oropharyngeal squamous cell carcinoma (**SCC**) * **cluster 1/2** (cancer cells) is surrounded by **cluster 0** (stroma) and **cluster 3** (immune cells) ![image-20201030113032968](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-033033.png) * **cluster1/2** presented a similar immune infiltration pattern, with an enrichment of proliferative **T-cells** and **SPP1 macrophages** * **cluster 3** presented a distinct immune infiltration pattern characterized by an enriched presence of (proliferative) **B-cells** * **cluster 0** harbored regulatory T-cells and terminally exhausted CD8 T-cells and was specifically enriched in **M2 macrophages** and **naive T-cells.** ![image-20201030132328526](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-052329.png) * Analysis of ductal breast carcinoma (**BC**) * also get a cancer-specific regional distribution: ![image-20201030134113951](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-054114.png) * subclonal was directly associated with local enrichment of distinct immune cell states ![image-20201030133721287](https://jieandze1314-1255603621.cos.ap-guangzhou.myqcloud.com/blog/2020-10-30-053721.png) * **Foresee** the regional distribution of immune cell types to become an important feature for the prediction of immuno-therapy outcome.