单细胞测序能区分“单个细胞基因表达”，却回答不了“基因表达的位置信息”——空间信息的缺失是其最大痛点。植物空间转录组虽好，却贵且难。我们能否另辟蹊径？例如下面的文章，没有做空间转录组但可以展示基因空间表达，利用的工具是ggplantmap这个R包：

水稻叶基因表达 （New Phytol. 2026 https://doi.org/10.1111/nph.71378）

下面介绍这个工具ggPlantmap：

他内部已经内置了很多植物组织图像，大家可以直接使用，如果想自己建立植物组织图像可以借助ICY软件和拍摄的图像手动绘制每个细胞的ROI生成组织图像，但这个非常费时费力，可以参考官方文档操作，文档地址： https://github.com/leonardojo/ggPlantmap/blob/main/guides/TutorialforXMLfile.pdf 如果不想自己费力建立组织图片，ggplantmap目前现有的植物组织图片数据可用如下：

ggPlantmap.name Species Tissue  Type  Description Layers  Image.Reference Made.by Contact.InfoggPm.At.roottip.crosssection  Arabidopsis thaliana  root  cross-section Cross-section of Arabidopsis thaliana root tips Cells https://www.future-science.com/doi/10.2144/000114621  Leonardo Jo l.jo@uu.nlggPm.At.roottip.longitudinal  Arabidopsis thaliana  root  longitudinal  Longitudinal view of Arabidopsis thaliana root tips Cells https://doi.org/10.1186/s13007-019-0417-9 Leonardo Jo l.jo@uu.nlggPm.At.3weekrosette.topview  Arabidopsis thaliana  rosette top view  Top view of 3.5 weel old Arabidopsis thaliana rosettes. Leaves  https://doi.org/10.1186/s12870-015-0483-8 Leonardo Jo l.jo@uu.nlggPm.At.leafepidermis.topview Arabidopsis thaliana  leaf epidermis  top view  Top view of Arabidopsis thaliana leaf adaxial epidermis Cells https://www.nature.com/articles/s41477-021-00906-0  Leonardo Jo l.jo@uu.nlggPm.At.leaf.crosssection Arabidopsis thaliana  leaves  cross-section Cross-section of Arabidopsis thaliana leaves  Cells https://doi.org/10.1199/tab.0181  Leonardo Jo l.jo@uu.nlggPm.At.seed.devseries  Arabidopsis thaliana  seed  development series  Diagram of the development of Arabidopsis thaliana seed tissues Cells,Tissues and Stage https://doi.org/10.1073/pnas.1222061110Leonardo Jo  l.jo@uu.nlggPm.At.earlyembryogenesis.devseries  Arabidopsis thaliana  embryo  development series  Diagram of the development of Arabidopsis thaliana embryos  Cells, Ontogeny and Stage https://doi.org/10.1111/nph.12267 Leonardo Jo l.jo@uu.nlggPm.At.shootapex.longitudinal  Arabidopsis thaliana  shoot apical meristem longitudinal  Diagram of the Arabidopsis thaliana shoot apical meristem Layers and Zone https://doi.org/10.1007/s10265-020-01174-3  Leonardo Jo l.jo@uu.nlggPm.At.inflorescencestem.crosssection  Arabidopsis thaliana  inflorescence stem  cross-section Cross-section of Arabidopsis thaliana inflorescence stem  Cells https://academic.oup.com/plcell/article/33/2/200/6017181  Leonardo Jo l.jo@uu.nlggPm.Sl.root.crosssection Solanum lycopersicum  root  cross-section Cross-section of Solanum lycopersicum (Tomato) root Cells https://doi.org/10.1016/j.cell.2021.04.024  Leonardo Jo l.jo@uu.nlggPm.At.leaf.topview  Arabidopsis thaliana  leaf  top view  Top view of Arabidopsis thaliana leaves from left to right in order of emergence from the SAM.  Leaves  http://doi.org/10.1199/tab.0181Leonardo Jo  l.jo@uu.nlggPm.At.rootelong.longitudinal  Arabidopsis thaliana  root elongation zone  longitudinal  Longitudinal view of Arabidopsis thaliana root at the elongation zone Cells https://doi.org/10.1016/j.devcel.2022.01.008  Leonardo Jo l.jo@uu.nlggPm.At.rootmatur.crosssection  Arabidopsis thaliana  root maturation zone  cross-section Cross-section of Arabidopsis thaliana root at the maturation zone Cells https://doi.org/10.1016/j.devcel.2022.01.008  Leonardo Jo l.jo@uu.nlggPm.At.flower.diagram  Arabidopsis thaliana  flower  diagram Diagram of the ABC model for Arabidopsis thaliana flowers Tissues and ABC.model Taiz, Lincoln, et al. Plant physiology and development. No. Ed. 6. Sinauer Associates Incorporated, 2015. Leonardo Jo l.jo@uu.nlggPm.At.lateralroot.devseries Arabidopsis thaliana  lateral root  development series  Diagram of the development of Arabidopsis thaliana lateral root Cells,Stage https://doi.org/10.3389/fpls.2019.00206 Leonardo Jo l.jo@uu.nlggPm.Ms.root.crosssection Medicago sativa root  cross-section Cross-section of Medicago sativa (Alfalfa) root Cells Unpublished Leonardo Jo l.jo@uu.nlggPm.Os.leaf.crosssection Oryza sativa  leaf  cross-section Cross-section of a large vascular bundle in the maturation zone of a developing rice leaf Cells manuscript in progress  Sean Robertson  rober136@myumanitoba.caggPm.At.seedling.saltdrought  Arabidopsis thaliana  seedling  diagram Diagram of Arabidopsis seedlings in control, salt (NaCl) and drought (Sorbitol) conditions  Tissues https://doi.org/10.64898/2025.12.03.691840  Kilian Duijts kilian.duijts@wur.nl

下面代码演示如何将单细胞数据基因表达数据在组织空间中展示：

1. 加载R包

# Loading Packageslibrary(ggPlantmap)library(tidyverse)

演示数据来源拟南芥根的单细胞转录组数据，展示四个基因（AT1G03550、AT1G09750、AT1G29950 和 AT5G06200）

Denyer, T., Ma, X., Klesen, S., Scacchi, E., Nieselt, K., & Timmermans, M. (2019). Spatiotemporal Developmental Trajectories in the Arabidopsis Root Revealed Using High-Throughput Single-Cell RNA Sequencing. Developmental cell, 48(6), 840–852.e5.

2. 单细胞转录组基因表达数据：

data <- read.csv("data/download_avg_UMI.csv")print(data)##           X   C0   C0.   C1   C1.   C2  C2.   C3   C3.   C4   C4.   C5   C5.## 1 AT1G03550 0.10 11.41 0.12 11.46 0.07 7.11 0.03  4.87 0.66 74.36 0.05  8.66## 2 AT1G09750 0.00  0.38 0.00  0.19 0.00 0.22 0.04  6.64 0.00  1.17 0.00  0.50## 3 AT1G29950 0.05  5.13 0.04  4.08 0.06 6.47 0.34 34.96 0.01  5.36 0.38 37.38## 4 AT5G06200 0.00  0.00 0.00  0.00 0.00 0.00 0.00  0.88 0.00  1.63 0.01  0.99##     C6   C6.   C7   C7.   C8   C8.   C9  C9.  C10 C10.  C11  C11.  C12  C12.## 1 0.04 10.03 0.02  7.46 0.04 11.51 0.02 3.23 0.02 5.20 0.01  2.74 0.10 10.49## 2 0.00  1.94 0.01  1.36 0.01  1.19 0.00 0.00 0.00 0.58 1.39 75.34 0.01  1.40## 3 0.02  8.74 0.03 10.85 0.01  3.97 0.03 5.91 0.04 8.67 0.00  0.68 0.26 23.78## 4 0.00  0.32 0.00  1.69 0.00  1.59 0.00 1.61 0.32 2.89 0.00  0.00 0.00  0.70##    C13  C13.  C14 C14.  C15  C15.  C16 C16.## 1 0.02  2.19 0.01  4.0 1.19 80.65 0.06 7.69## 2 0.00  0.00 0.00  0.8 0.00  0.00 0.00 0.00## 3 1.48 71.53 0.03  6.4 0.00  0.00 0.03 5.13## 4 0.00  0.00 0.00  1.6 0.00  0.00 0.00 0.00

3. 数据处理：

数据以基因名为第一列，后为特定簇（C0 至 C16）中所有细胞的平均表达（原始 UMI 读数）。注意每个簇都有一列额外列（例如：C0.），描述簇中在某一阈值下表达该基因的细胞百分比。目前，我们只关注平均表达式值。 将数据转换成长型：

processed.data <- data %>%    pivot_longer(-X)%>%## making it tidy    filter(!str_detect(name,"[.]"))## removing columns that correspond to the percentage of cells with signalprocessed.data## # A tibble: 68 × 3##    X         name  value##    <chr>     <chr> <dbl>##  1 AT1G03550 C0     0.1 ##  2 AT1G03550 C1     0.12##  3 AT1G03550 C2     0.07##  4 AT1G03550 C3     0.03##  5 AT1G03550 C4     0.66##  6 AT1G03550 C5     0.05##  7 AT1G03550 C6     0.04##  8 AT1G03550 C7     0.02##  9 AT1G03550 C8     0.04## 10 AT1G03550 C9     0.02## # ℹ 58 more rows

现在我们有基因平均表达式值。但是，为了将它们映射到 ggPlantmap，每个cluster与的细胞类型对应关系需要对应起来：

reference <- read.table("data/Summary of celltypes Arabidopsis Root.txt",fill=T,sep="\t",header=T)reference##    Cluster         Identity## 1       C0 Lateral Root Cap## 2       C1 Lateral Root Cap## 3       C2 Lateral Root Cap## 4       C3         Meristem## 5       C4      Trichoblast## 6       C5           Phloem## 7       C5        Pericycle## 8       C6     Atrichoblast## 9       C7         Meristem## 10      C8         Meristem## 11      C9         Meristem## 12     C10       Endodermis## 13     C11           Cortex## 14     C12        Columella## 15     C13            Xylem## 16     C14          Unknown## 17     C15      Trichoblast## 18     C16 Lateral Root Cap

ggplantmap 根横切面示意图数据：

ggPm.At.rootmatur.crosssection## # A tibble: 1,113 × 5##    ROI.name     ROI.id point     x     y##    <chr>         <int> <int> <dbl> <dbl>##  1 Atrichoblast      1     1  968. -323.##  2 Atrichoblast      1     2  987. -305.##  3 Atrichoblast      1     3  999. -300.##  4 Atrichoblast      1     4 1019. -320.##  5 Atrichoblast      1     5 1027. -332.##  6 Atrichoblast      1     6 1024. -353.##  7 Atrichoblast      1     7 1012. -366.##  8 Atrichoblast      1     8  994. -369.##  9 Atrichoblast      1     9  980. -367.## 10 Atrichoblast      1    10  972. -365.## # ℹ 1,103 more rowsggPlantmap.plot(ggPm.At.rootmatur.crosssection)

拟南芥根横切面图

4. 创建空间表达图

以上数据已经都准备好了，下面的示例展示了一个 for 循环，为 processed.data 对象中每个基因生成一个基因空间表达图：

for  (k in unique(data$X)){  single.data <- processed.data %>%    filter(X == k)%>%## filtering for one gene    merge(reference,by.x="name",by.y="Cluster",all.x=T)## combining with the reference dataset for labeling clusters  ## Merging with the ggPm.At.rootmatur.crossection using the ggPlantmap.merge() function  final.table <- ggPlantmap.merge(ggPm.At.rootmatur.crosssection,single.data,id.x ="ROI.name",id.y="Identity")  ## Using the ggPlantmap.heatmap to create a heatmapplot <- ggPlantmap.heatmap(final.table,value.quant = value)+     scale_fill_gradient(low="white",high="red",limits=c(0,1.5))+    labs(title=paste0(k))## Title of the plot should be name of the gene  print(plot)## printing the plot  ##ggsave(plot=plot,paste0(k,".png"),dpi=300) ## in case you want to save your heatmap}

单细胞转录组基因空间表达热图数据和示例代码我已经放到我们云服务器中了，大家按照下面的链接领服务器获取代码和练习绘制：生信课堂单细胞转录组专用云服务器免费领取更多植物单细胞转录组文献解读：Genome Biology|8个单细胞+1个空转葡萄叶片生物胁迫Horticulture Research|2个单细胞转录组测序与空间代谢组学分析揭示银杏外种皮PBJ｜1个样本单细胞发一区文章-蛇根草喜树碱生物合成中的细胞特异性表达研究Nature Plants｜7个样本单细胞转录组-拟南芥种子早期发育图谱Horticulture Research｜4个样本单细胞转录组发一区文章-香蕉根枯萎病PBJ | 12个样本单细胞转录组发高分文章=番茄抗感品种抗ToBRFV病毒差异Nature Communications｜4个样本单细胞转录组发高分文章-玉米根热胁迫New Phytologist｜11个水稻叶片胁迫单细胞转录组-几乎没做实验就发了1区文章如何做到的