rnadtu 0.1.8

Package for running the CIDR algorithm implemented in R, in python

rnadtu

A Python package for running the CIDR algorithm (originally implemented in R) for clustering single-cell RNA-seq data using AnnData objects.

Features

• Run the CIDR algorithm from Python using:

  • cidr() — subprocess-based, using temporary CSV files (recommended for large datasets)
  • cidr_non_csv() — subprocess-based, using in-memory buffers (faster but memory intensive and difficult to debug)
  • cidr_rpy2() — uses the rpy2 bridge (direct R integration, less suitable for large datasets)

• Optional configuration options:

  • data_type: default is "raw", can be "cpm" (counts per million)
  • n_cluster: default is None (CIDR calculates the optimal number), or manually specify an positive integer
  • layer: which layer of the AnnData object to use for input
  • Optional output controls:
    • pc: store principal coordinates
    • dissim: store dissimilarity matrix
    • dropout: store dropout matrix
    • save_clusters: (only in cidr_rpy2) choose store cluster labels (default True)

• Results are saved in the AnnData object:

  • obsm[layer + "_cidr_pc"] — principal components
  • obsm[layer + "_cidr_clusters"] — cluster labels (if save_clusters=True)
  • obsp[layer + "_cidr_dissimilarity_matrix"] — pairwise distances
  • uns[layer + "_cidr_variation"], uns[layer + "_cidr_eigenvalues"] — PCA variation
  • uns[layer + "_cidr_dropout_candidates"] — dropout data

• Generates a clustering plot in cidr_plots.pdf

Installation

Important: Before installing the package via pip install rnadtu, please follow the installation prerequisites below

Install R and CIDR

To get the package up and running you first need to install R and the CIDR library:

• Install R ≥ 4.4.0 with default settings
• On windows, make sure R is added to PATH, typically as
  C:\Program Files\R\R-<your-version>\bin
• Install Corresponding version of RTools
• Open an R console in a terminal or in Rstudio and run the following command:
  install.packages("devtools") (this may take a little)
• install CIDR with the following command:
  devtools::install_github('VCCRI/CIDR')
• OPTIONAL if you want to run individual functions of the CIDR algorithm:
  • install the Arrow with the following command:
    install.packages('arrow')
  • install the qs package with the following command:
    install.packages('qs')

---

Install rpy2

To install rpy2, which is a package that some of the functions in our package depends on, simply use the command pip install rpy2. However, if you are on windows, do the following:
The R-version in the steps is 4.4.0, but should be replaced with your respective version, this should be replaced with the R-version on your pc.

• Make sure python is installed with Python --version
• Update your pip version with python -m pip install --upgrade pip
• Make sure you have R installed, typically placed at
  C:\Program Files\R\R-4.4.0
• As mentioned earlier, add the following to your PATH variable
  C:\Program Files\R\R-4.4.0\bin (No action needed for mac)
• Install rpy2 without compiling with
  pip install --only-binary :all: rpy2(Skip this step for mac)

Finally, go ahead and install the rnadtu package with
pip install rnadtu

License

RNADTU is licensed under the GNU General Public License v2.0.

It includes source code from the VCCRI/CIDR project, which is also distributed under the GPL v2.0.

See the License file for full details.