cellchat netEmbedding 运行出错

运行cellchat <- netEmbedding(cellchat,type = “functional”) 报错如下：

Error in runUMAP(Similarity, min.dist = 0.3, n.neighbors = k) : Cannot find UMAP, please install through pip (e.g. pip install umap-learn or reticulate::py_install(packages = 'umap-learn')).

因为没有安装umap实现，按提示在终端使用pip install umap-learn，直接运行后报错还是一样，因为你没有指定R使用的python版本：

重启Rstudio后：

library(reticulate)
use_python('/usr/bin/python3',required = T) #指定安装umap-learn那个版本的python位置
py_version() #看看是不是指定正确

最后运行后报错：

>cellchat <- netEmbedding(cellchat, type = "functional") 
 Manifold learning of the signaling networks for a single dataset 
/Users/yangl/Library/Python/3.8/lib/python/site-packages/umap/umap_.py:132: UserWarning: A large number of your vertices were disconnected from the manifold. 
Disconnection_distance = 1 has removed 142 edges. 
It has fully disconnected 3 vertices. You might consider using find_disconnected_points() to find and remove these points from your data. 
Use umap.utils.disconnected_vertices() to identify them. 
warn(

或者直接通过运行：reticulate::py_install(packages = ‘umap-learn’)。但也依然报错：

>cellchat <- netEmbedding(cellchat, type = "functional")
Manifold learning of the signaling networks for a single dataset 
C:\Users\zzu\AppData\Local\R-MINI~1\envs\R-RETI~1\lib\site-packages\umap\umap_.py:133: UserWarning: A large number of your vertices were disconnected from the manifold.
Disconnection_distance = 1 has removed 142 edges.
It has fully disconnected 3 vertices.
You might consider using find_disconnected_points() to find and remove these points from your data.
Use umap.utils.disconnected_vertices() to identify them.
f"A large number of your vertices were disconnected from the manifold.\n"

通过google搜索find_disconnected_points(),GitHub issue中有人提问并得到解决方法：https://github.com/sqjin/CellChat/issues/167（即 source(file = ‘CellChat_issue167_netClusteringFix.R’) #使用外部导入修改的函数），但依然报错如旧。

最后我通过查看GitHub issue中所有与netEmbedding函数相关的issue，最终找到了解决办法：https://github.com/sqjin/CellChat/issues/196 （换个umap算法实现,结果也会有些许差异）

install.packages("uwot")
library(uwot)
cellchat <- netEmbedding(cellchat, umap.method = 'uwot',type = "functional") #调用uwot实现umap, 程序自带，只需使用参数umap.method = 'uwot'，无需修改对应的netEmbedding函数源码

尊重他人劳动成果，转载请注明出处:Bluesky's blog » cellchat netEmbedding 运行出错