Question: Can I target higher cell numbers when performing Single Cell 5’ CRISPR experiments?
Answer: The multiplet rate for Chromium products is determined by Poisson loading of cells into GEMs that contain barcoded gel beads. As the targeted cell recovery increases, the percentage of GEMs that contain multiple cells increases (see multiplet rate table under Tips & Best Practices - Cell Concentration in the User Guide CG000736).
For standard operation of our single cell GEM-X Universal Gene Expression assays with a single sample per lane, we recommend loading cells at a concentration designed to recover a maximum of 20,000 productive cell barcodes, which results in a multiplet rate of ~8%.
It is possible to load higher cell numbers onto the chip, but when doing so, the multiplet rate increases. Many users routinely load higher cell numbers when they use multiplexing strategies such as genetic demultiplexing or various “hashing” approaches that allow them to discern which cell barcodes contain cell multiplets.
On top of gene expression, surface protein detection and immune receptor profiling, our single cell 5’ product can also be used to directly capture sgRNAs used for CRISPR/Cas9-mediated perturbations including small and large scale CRISPR screens. In this application, researchers need to reliably identify the perturbation in the cell and couple it to the gene expression phenotype. The high expression of the sgRNA coupled with the high efficiency of direct guide capture in this system results in high confidence assignment of perturbations to cells.
In the context of CRISPR screens, the sgRNA can act as a multiplexing barcode for purposes of multiplet discrimination: cells that contain multiple guides can be flagged as multiplets and discarded†. This “internal hashing” property of single cell CRISPR screens allows customers to load much higher numbers of cells per lane into 10x 5’ v3 on GEM-X and confidently recover more single cells without any additional workflow steps.
As an example, loading cells at a concentration to recover ~60,000 cell barcodes from a single lane of GEM-X will result in ~45,000 singlets and ~15,000 multiplets. Using this strategy can greatly increase the throughput of single cell CRISPR experiments and reduce cost to the user.
Below is a Cell Multiplet Table for GEM-X Universal 5’ Singleplex with approximate multiplet rate, cells loaded, cells recovered, and the breakdown of singlets and multiplets.
Users need to balance the cost savings obtained from generating more singlets per lane with the cost of wasted sequencing of multiplets, but most find that overloading as described above results in favorable economics.
Please note that the demultiplexing using sgRNA is not supported by 10x Genomics.
Please refer to the following article to determine what assay best suits your experimental goals:
Which 10x Genomics Single Cell CRISPR approach should I use?
† There are some “compressed sensing” style screens where researchers deliberately generate a high rate of multiplets, either through high MOI transductions, or deliberate generation of cell multiplets. In these cases, doublets are kept and phenotypes determined by deconvolution strategies.
Product: Single Cell Immune Profiling