In satijalab/seurat-wrappers: Community-Provided Methods and Extensions for the Seurat Object

This vignette demonstrates how to run PaCMAP, a dimensionality reduction method that can be used for providing robust and trustworthy visualization, on a Seurat object. If you use our work, please cite both papers:

Understanding How Dimension Reduction Tools Work: An Empirical Approach to Deciphering t-SNE, UMAP, TriMap, and PaCMAP for Data Visualization

Yingfan Wang, Haiyang Huang, Cynthia Rudin & Yaron Shaposhnik

Journal of Machine Learning Research, 2021

doi: https://doi.org/10.48550/arXiv.2012.04456

Towards a comprehensive evaluation of dimension reduction methods for transcriptomic data visualization

Haiyang Huang, Yingfan Wang, Cynthia Rudin and Edward P. Browne

Communications biology, 2022

doi: https://doi.org/10.1038/s42003-022-03628-x

GitHub: https://github.com/YingfanWang/PaCMAP

knitr::opts_chunk$set(
  message = FALSE,
  warning = FALSE,
  fig.width = 10
)

Prerequisites to install:

• Seurat
• SeuratWrappers
• SeuratData
• Reticulate

In addition to R packages, PaCMAP relies on Python to deliver high performance. To streamline the installation process and make environment management easier, we strongly recommend you to use anaconda(https://www.anaconda.com/download) or miniconda(https://docs.anaconda.com/miniconda/miniconda-install/) for managing Python environments. Below, we provide step-by-step instructions on how to properly install PaCMAP after you have installed one of these tools.

Create a conda environment with PaCMAP installed:

```{bash conda environment setup, eval=FALSE} conda create -n "pacmap" python=3.12 # Install in the environment called "pacmap" conda activate pacmap conda install -y conda-forge::pacmap

To run PaCMAP, you need to connect your R console to the corresponding conda environment.
If your Conda/Miniconda installation is located in a non-default directory, you might
set up the conda variable as `/path/to/your/conda`. This ensures the correct environment
is used during the installation.

```r
reticulate::use_condaenv(condaenv = "pacmap", conda = "auto")

library(Seurat)
library(SeuratData)
library(SeuratWrappers)

PaCMAP on PBMC3k

To learn more about this dataset, type ?pbmc3k

InstallData("pbmc3k")
pbmc3k.final <- LoadData("pbmc3k",type="pbmc3k.final")

# Initial processing to select variable features
pbmc3k.final <- UpdateSeuratObject(pbmc3k.final)
pbmc3k.final <- FindVariableFeatures(pbmc3k.final)

# run PaCMAP on Seurat object. 
pbmc3k.final <- RunPaCMAP(object = pbmc3k.final, features=VariableFeatures(pbmc3k.final))

# visualize markers
features.plot <- c('CD3D', 'MS4A1', 'CD8A', 'GZMK', 'GZMB', 'FCGR3A')
DimPlot(object=pbmc3k.final,reduction="pacmap")

pbmc3k.final <- NormalizeData(pbmc3k.final, verbose = FALSE) 
FeaturePlot(pbmc3k.final, features.plot, ncol = 2, reduction="pacmap")

You can also specify dims of your original dataset for running PaCMAP

# run PaCMAP on Seurat object. 
pbmc3k.final <- RunPaCMAP(object = pbmc3k.final, dims=2:5)
# visualize markers
features.plot <- c('CD3D', 'MS4A1', 'CD8A', 'GZMK', 'GZMB', 'FCGR3A')
DimPlot(object=pbmc3k.final,reduction="pacmap")

satijalab/seurat-wrappers documentation built on Nov. 23, 2024, 5:09 p.m.