README

中文版这里

This document describes how to perform file read and write operations in R and Python. In both R and Python, read and write operations are straightforward and can be accomplished with just one line of code.

Install easySCF

To install easySCF, you can use the following code:

# Install easySCF in R
install.packages("devtools")
devtools::install_github("xleizi/easySCF/r")

# Install easySCF in Python
!pip install easySCFpy

Read and Write in R

Reading in R

In R, you can use the following function for file read and write operations:

• loadH5(): Used to read H5 files.

Parameter explanation:

• FileName: String, required. Specifies the path to the HDF5 file to be loaded.
• assay: String, default is "RNA". Specifies the type of layers to access in the HDF5 file.
• SeuratVersion: Provided by default by checkSeuratVersion(). This determines the version of the Seurat object used when loading data.
• image_name: String, default is "Spatial". Used to identify the name of the image data (if any).
• useBPcells: Boolean, default is FALSE. Whether to use BPcells for data storage.
• useDisk: Boolean, default is TRUE. Whether to use disk storage to optimize memory usage for BPcells data.
• calData: Boolean, default is TRUE. Whether to calculate normalization after loading data.
• calScale: Boolean, default is FALSE. Whether to scale the data.
• calFeatures: Boolean, default is FALSE. Whether to calculate highly variable genes.
• group_by: Default is NULL. Used to specify grouping variables, commonly used in subsequent analysis to differentiate between different data subsets.
• readType: String, default is "Seurat". Specifies the method or format of data reading, usually corresponding to the analysis software.

Here is a simple example demonstrating how to use these functions:

library(easySCFr)
sce <- loadH5("data.h5")

Writing in R

In R, you can use the following function for file read and write operations:

• saveH5(): Used to save Seurat objects to an H5 file.

Parameter explanation:

• data: Required. The Seurat object to be saved.
• FileName: String, required. Specifies the path and name of the file to save.
• assay: String, default is "RNA". Specifies the type of layers data contained in the file.
• save_graph: Boolean, default is TRUE. Whether to save data on cell interconnections.
• SeuratVersion: Provided by default by checkSeuratVersion(). This determines the version of the Seurat object used when saving data.
• image_name: String, default is NULL. If provided, specifies the name of the image file associated with the data.
• split_save: Boolean, default is TRUE. Whether to split the data into multiple subsets for saving, suitable for very large data sets.
• max_cells_per_subset: Integer, default is 5000. Specifies the maximum number of cells per subset during split saving.

Here is a simple example demonstrating how to use these functions:

library(easySCFr)
saveH5(sce, "data.h5")

Read and Write in Python

Reading in Python

In Python, you can use the following function for file read and write operations:

• loadH5(): Used to read H5 files.

Parameter explanation:

• filename: str | Path, required. Specifies the path to the HDF5 file to be loaded.
• assay: str, default is "RNA". Specifies the type of layers data to be read from the HDF5 file.
• datatype: str, default is "scanpy". Specifies the data format, can be saved in different data types.
• image_name: str | None, optional. If provided, specifies the name of the image file associated with the data.
• backed: bool | Literal['r', 'r+'] | None, optional. Specifies the file read mode, 'r' for read-only mode, 'r+' for read-write mode, None for loading all into memory.
• as_sparse: Sequence[str], default is "raw/X". Specifies which data should be stored in a sparse matrix format.
• as_sparse_fmt: type[spmatrix], default is sparse.csr_matrix. Defines the format of the saved sparse matrix, modification is not recommended.

Here is a simple example demonstrating how to use these functions:

from easySCFpy import loadH5
sce = loadH5("data.h5")

Writing in Python

In Python, you can use the following function for file read and write operations:

• saveH5(): Used to save Seurat objects to an H5 file.

Parameter explanation:

• adata: Any, required. The AnnData object to be saved.
• h5_path: Path | str, required. The path for saving the HDF5 file.
• assay: str, default is "RNA". Specifies the type of layers data.
• datatype: str, default is "scanpy". Specifies the data format.
• image_name: str, default is "slice". Specifies the name of the image file associated with the data.
• save_graph: bool, default is True. Whether to save data on cell interconnections.
• as_dense: Sequence[str], default is an empty tuple. Manually specifies which data fields should be saved in a dense matrix format.
• split_save: bool, default is True. Whether to split the data into multiple subsets for saving, usually used for very large data sets.
• max_cells_per_subset: int, default is 5000. Specifies the maximum number of cells per subset.
• compression: Literal['gzip', 'lzf'] | None, default is "gzip". Specifies the compression algorithm used.
• compression_opts: int | None, optional. Provides specific parameters for the compression algorithm.

Here is a simple example demonstrating how to use these functions:

from easySCFpy import saveH5
saveH5(sce, "data.h5")