Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
AlicenJoyHenning authored Sep 22, 2024
1 parent 7c305ae commit 3e2a9ad
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ Single-cell RNA sequencing (scRNA-seq) is a well-established technique in the er

<br><br>

For droplet-based scRNA-seq protocols, ‘low quality’ cells are those that originate from droplets that contain either more than one cell (doublet), no cells (empty), or a damaged cell. While many quality control tools available are designed to identify doublets and empty droplets, few are specialised in identifing damaged cells. Damaged cell detection is more often achieved by setting thresholds for metrics such as average mitochondrial gene expression or UMI and features counts per barcode. These thresholds, even when dynamically calculated, vary across samples, cell types, tissues, treatment conditions, and species. This could result in overly stringent filtering, where many true cells are excluded from downstream but also in overly conservative filtering where many contaminating damaged cells remain. But searching for true damaged droplets in a sample-specific manner is tedious and not always intuitive, leading to user-defined filtering that lacks reproducibility.
For droplet-based protocols, ‘low quality’ cells are those that originate from droplets that contain either more than one cell (doublet), no cells (empty), or a damaged cell. While many quality control tools available are designed to identify doublets and empty droplets, few are specialised in identifing damaged cells. Damaged cell detection is more often achieved by setting thresholds for metrics such as average mitochondrial gene expression or UMI and features counts per barcode. These thresholds, even when dynamically calculated, vary across samples, cell types, tissues, treatment conditions, and species. This could result in overly stringent filtering, where many true cells are excluded from downstream but also in overly conservative filtering where many contaminating damaged cells remain. But searching for true damaged droplets in a sample-specific manner is tedious and not always intuitive, leading to user-defined filtering that lacks reproducibility.

<br>

Expand Down

0 comments on commit 3e2a9ad

Please sign in to comment.