从命令行使用scanpy的脚本
scanpy-scripts的Python项目详细描述
稀疏脚本
从命令行使用scanpy的脚本
为了用任何给定的工作流语言包装scanpy的内部工作流,让脚本调用其中的每个步骤是很重要的。这些脚本正在这里编写,随着时间的推移,它们的完整性将得到提高。
安装
conda install scanpy-scripts
# or
pip3 install scanpy-scripts
测试安装
其中包括一个示例脚本:
scanpy-scripts-tests.sh
这将下载a well-known test 10X dataset并执行下面描述的所有脚本。
命令
可用命令如下所述。每个都有通过--help获得的使用说明,有关更多详细信息,请参阅scanpy中的函数文档。
Usage: scanpy-cli [OPTIONS] COMMAND [ARGS]...
Command line interface to [scanpy](https://github.com/theislab/scanpy)
Options:
--debug Print debug information
--verbosity INTEGER Set scanpy verbosity
--version Show the version and exit.
--help Show this message and exit.
Commands:
read Read 10x data and save in specified format.
filter Filter data based on specified conditions.
norm Normalise data per cell.
hvg Find highly variable genes.
scale Scale data per gene.
regress Regress-out observation variables.
pca Dimensionality reduction by PCA.
neighbor Compute a neighbourhood graph of observations.
embed Embed cells into two-dimensional space.
cluster Cluster cells into sub-populations.
diffexp Find markers for each clusters.
paga Trajectory inference by abstract graph analysis.
dpt Calculate diffusion pseudotime relative to the root cells.
plot Visualise data.