scanpy-GPU#
These functions offer accelerated near drop-in replacements for common tools provided by scanpy
.
Preprocessing pp
#
Filtering of highly-variable genes, batch-effect correction, per-cell normalization.
Any transformation of the data matrix that is not a tool. Other than tools
, preprocessing steps usually don’t return an easily interpretable annotation, but perform a basic transformation on the data matrix.
Basic Preprocessing#
|
Calculates basic qc Parameters. |
|
Filter cell outliers based on counts and numbers of genes expressed. |
|
Filter genes based on number of cells or counts. |
|
Normalizes rows in matrix so they sum to |
|
Calculated the natural logarithm of one plus the sparse matrix. |
|
Annotate highly variable genes. |
|
Use linear regression to adjust for the effects of unwanted noise and variation. |
|
Scales matrix to unit variance and clips values |
|
Performs PCA using the cuml decomposition function. |
|
Applies analytic Pearson residual normalization, based on Lause21. |
|
Flags a gene or gene_family in .var with boolean. |
|
Filters the |
Batch effect correction#
|
Use harmonypy to integrate different experiments. |
Doublet detection#
|
Predict doublets using Scrublet. |
|
Simulate doublets by adding the counts of random observed transcriptome pairs. |
Neighbors#
|
Compute a neighborhood graph of observations with cuml. |
Tools: tl
#
tools
offers tools for the accelerated processing of AnnData
. For visualization use scanpy.pl
.
Embedding#
|
Embed the neighborhood graph using UMAP's cuml implementation. |
|
Performs t-distributed stochastic neighborhood embedding (tSNE) using cuml library. |
|
Diffusion maps has been proposed for visualizing single-cell data. |
|
Force-directed graph drawing with cugraph's implementation of Force Atlas 2. |
|
Util to run |
|
Calculate the density of cells in an embedding (per condition). |
Clustering#
|
Performs Louvain clustering using cuGraph, which implements the method described in: |
|
Performs Leiden clustering using cuGraph, which implements the method described in: |
Marker genes#
|
Rank genes for characterizing groups. |
Plotting#
For plotting please use scanpy’s plotting API scanpy.pl
.