Question: What metrics should I be looking at to optimize my sample prep in my pilot experiment?
Answer : When optimizing sample preparation, a pilot experiment comparing two or more conditions may be useful. The Single Cell Gene Expression LT kit may be useful to provide sequencing readouts for pilot experiments.
At the sample stage:
A high-quality cell suspension has high cell viability (>90% ideally, >70% is accepatable), low cell debris and low cell clumping. A high-quality nuclei suspension has low cell viability (<0.5%, as nuclei stain dead), low cellular debris, low nuclei clumping and high-quality nuclear membranes. As you evaluate different conditions, it may be useful to compare the following metrics at the sample QC stage:
- Cell viability
- Cell or nuclei yield
- Cell/nuclei clumping
- Nuclei quality, see: What are the best practices for working with nuclei samples for 3' single-cell gene expression?
In the resulting data:
The web summary file and cloupe file can be useful to evaluate and compare data between different conditions. See Interpreting Cell Ranger Web Summary Files for Single Cell Gene Expression Assays for an explanation of different summary metrics. For data comparison across samples, follow the “How to Compare Results from Different Samples” section in Tutorial: Analyze scRNA-Seq Data From a Publication Using 10x Software. The following metrics may be useful for comparison:
- Fraction of reads in cells: Can be used to assess ambient RNA, see: How to interpret the "Fraction Reads in Cells" metric?
- Cluster structure: Does one condition give better separation than another?
- Differentially expressed genes: Look out for upregulation of stress genes or mitochondrial/ribosomal genes, see: Why do I see a high level of mitochondrial gene expression?
- Cell types of interest: Is your cell type of interest better preserved in one condition vs the other?
Taking all these pieces of information into consideration may help you determine between two sample preparation conditions before scaling up your experiments.
Products : Single Cell Gene Expression