isolate umap from scanpy to scv

elenas club

2 min read

·

13-02-2025

10

0

Share

This article details how to extract UMAP coordinates generated by Scanpy and save them into a simple CSV file for further analysis or visualization in other tools. UMAP (Uniform Manifold Approximation and Projection) is a powerful dimensionality reduction technique frequently used in single-cell RNA sequencing (scRNA-seq) analysis to visualize high-dimensional data. Scanpy is a popular Python library for analyzing scRNA-seq data, and often the first step in visualizing your data is using UMAP. This guide will show you how to easily export those coordinates.

Setting up your environment

Before we begin, ensure you have the necessary libraries installed. You'll need Scanpy and Pandas:

pip install scanpy pandas

We'll also assume you have already pre-processed your scRNA-seq data and performed UMAP dimensionality reduction within Scanpy. If not, refer to the Scanpy documentation for guidance on data preprocessing and UMAP integration.

Extracting UMAP Coordinates

Let's say your Scanpy AnnData object is called adata. The UMAP coordinates are stored within the .obsm attribute. Specifically, you'll find them under the key 'X_umap'.

import scanpy as sc
import pandas as pd

# Load your AnnData object
adata = sc.read_h5ad("your_adata_file.h5ad") # Replace with your file

# Extract UMAP coordinates
umap_coords = adata.obsm['X_umap']

# Check the shape of the coordinates.  This should be (number of cells, 2)
print(umap_coords.shape)

Creating a Pandas DataFrame

To easily export to CSV, let's convert the NumPy array into a Pandas DataFrame:

# Create a Pandas DataFrame
df_umap = pd.DataFrame(umap_coords, columns=['UMAP1', 'UMAP2'])

#Optionally add cell barcodes as index:

df_umap.index = adata.obs.index

Saving to CSV

Finally, save the DataFrame to a CSV file:

# Save the DataFrame to a CSV file
df_umap.to_csv("umap_coordinates.csv")

This will create a file named umap_coordinates.csv containing two columns: UMAP1 and UMAP2, representing the x and y coordinates of each cell in the UMAP embedding. If you added the index, the first column will be your cell barcodes. This CSV file can then be imported into various tools like spreadsheets or visualization software.

Troubleshooting and Potential Issues

• KeyError: If you encounter a KeyError: 'X_umap', it means that UMAP hasn't been run on your adata object yet. You need to run the sc.tl.umap() function before extracting the coordinates. For example: sc.tl.umap(adata)

• Incorrect Shape: If the shape of umap_coords isn't (number of cells, 2), there might be an issue with your UMAP computation. Double-check your Scanpy code and parameters.

• Large Datasets: For extremely large datasets, consider using chunking or other memory-efficient techniques to avoid memory errors.

This method provides a straightforward and efficient way to isolate and save your UMAP coordinates for further analysis or visualization outside of the Scanpy environment. Remember to adapt file paths and variable names to your specific project setup. This allows flexible integration with other bioinformatics tools and workflows.