Understanding Quality Control in Single-Cell RNA Sequencing: Part II – Detecting Empty Droplets

July 4, 2024

In our first blog post, we discussed the importance of detecting and filtering out low UMI cells to ensure high-quality single-cell RNA sequencing (scRNA-seq) data. In this second part of our series on scRNA-seq quality control (QC), we will focus on detecting empty droplets. We’ll use the πšœπš’πš—πšπš•πšŽπ™²πšŽπš•πš•πšƒπ™Ί toolkit to illustrate this process.

What Are Empty Droplets?

In droplet-based scRNA-seq platforms, such as 10x Genomics, individual cells are encapsulated in tiny droplets along with barcoded beads. Ideally, each droplet should contain a single cell. However, many droplets end up empty or containing ambient RNA. These empty droplets, if not identified and removed, can introduce noise and distort downstream analyses.

Why Detect Empty Droplets?

Empty droplets typically contain very few RNA molecules, mainly ambient RNA that contaminates the solution. Including these in the analysis can lead to false positives and obscure true biological signals. Detecting and removing empty droplets is crucial to maintaining data integrity.

Step-by-Step Guide to Detecting Empty Droplets with πšœπš’πš—πšπš•πšŽπ™²πšŽπš•πš•πšƒπ™Ί

πšœπš’πš—πšπš•πšŽπ™²πšŽπš•πš•πšƒπ™Ί provides a straightforward approach to identifying empty droplets. Here’s how to do it:

Step 1: Load the Data

First, load your scRNA-seq data into R. πšœπš’πš—πšπš•πšŽπ™²πšŽπš•πš•πšƒπ™Ί supports various data formats, including πš‚πš’πš—πšπš•πšŽπ™²πšŽπš•πš•π™΄πš‘πš™πšŽπš›πš’πš–πšŽπš—πš objects and πš‚πšŽπšžπš›πšŠπš objects.

Step 2: Detect Empty Droplets

Use the πš›πšžπš—π™΄πš–πš™πšπš’π™³πš›πš˜πš™πšœ function, which implements the EmptyDrops method by Sun AT, et al. 2019. This method uses a statistical test to differentiate between real cells and empty droplets based on the total UMI count per droplet.

Step 3: Examine the Results

The πš›πšžπš—π™΄πš–πš™πšπš’π™³πš›πš˜πš™πšœ function adds columns to the cell metadata indicating the likelihood that a droplet contains a cell versus being empty.

Step 4: Filter Out Empty Droplets

You can filter out droplets that are likely empty by using the provided thresholds or by examining the distribution of the FDR values.

Practical Example

Let’s apply this to the πšœπšŒπ™΄πš‘πšŠπš–πš™πš•πšŽ dataset included in πšœπš’πš—πšπš•πšŽπ™²πšŽπš•πš•πšƒπ™Ί.

In this example, droplets with FDR values below 0.01 are identified as empty and removed from the dataset, ensuring that only droplets containing real cells are retained.

The resulting figure highlights the identified empty droplets.

Conclusion

Detecting and filtering out empty droplets is an essential step in scRNA-seq quality control. By using πšœπš’πš—πšπš•πšŽπ™²πšŽπš•πš•πšƒπ™Ί, researchers can efficiently identify these droplets, reducing noise and improving the accuracy of their analyses. In the next part of this series, we will explore methods to detect doubletsβ€”droplets that contain more than one cell. Stay tuned!


By following these guidelines, you can enhance the quality of your scRNA-seq data, paving the way for more reliable and insightful biological discoveries.

References

Hong R, Koga Y, Bandyadka S, Leshchyk A, Wang Y, Akavoor V, Cao X, Sarfraz I, Wang Z, Alabdullatif S, Jansen F. Comprehensive generation, visualization, and reporting of quality control metrics for single-cell RNA sequencing data. Nature communications. 2022 Mar 30;13(1):1688.

Lun AT, Riesenfeld S, Andrews T, Dao TP, Gomes T, Participants in the 1st Human Cell Atlas Jamboree, Marioni JC. EmptyDrops: distinguishing cells from empty droplets in droplet-based single-cell RNA sequencing data. Genome biology. 2019 Dec;20:1-9.

Schedule your free discovery call here

Contact us atΒ info@insigene.com

Β© 2024 INSiGENe Ltd. Site maintained by NFIC ServicesΒ  |Β  Privacy Policy.