Very excited to share our new preprint describing a method for ultra-high throughput single-cell RNA-seq biorxiv.org/content/10.110… #scRNAseq 
Our new method (scifi-RNA-seq) uses a mixture of plate-based combinatorial and droplet-based fluidic indexing to achieve an unprecedented balance of ease-of-use and throughput in a single experiment.
In droplet-based devices, generation of successful events requires a single bead and a single cell to be packaged in a droplet - each can be see as a Poisson process. To assure each droplet has only one of each, the rates of both must be tightly controlled.
Due to the usage of deformable beads in some platforms, bead loading is fairly efficient, so we postulated that it would be possible to use the full potential of the device if we would pre-label nuclei using the combinatorial indexing approach and load the device much more.
We wanted to stick with a widely-used, commercially available platform, so we started by trying to understand the nuclei and bead loading processes in the @10xGenomics Chromium device.
In a simple but elegant experiment, @PaulDatlinger loaded the 10x Chromium device (no lysis buffer) with up to 100 times more nuclei (1.5 million) and counted the number of nuclei in each droplet under the microscope.
By probabilistic modeling of the observed distributions of nuclei per droplet we were able to assess the linear range of the device and make predictions on the fraction of expected collisions with and without pre-labeling of cells. More on that here: andre-rendeiro.com/2019/12/13/chr…
We confirmed the computational predictions experimentally with Barnyard experiments and profiled primary human cells and cell lines, where we got up to 150,000 cells from one Chromium channel
Yes - this means the little black box your lab *already owns* can produce ~15X more output per channel. Using the 8 channels of the device one could profile 1 million cells in one go.
Since scifi-RNA-seq effectively removes the high cost of profiling large numbers of single cells (seq is still a bottleneck tho), we anticipate scifi-RNA-seq to be of use for organism-scale profiling efforts and massive-scale perturbation studies such as CROP-seq or drug screens
As always it has been a great privilege and pleasure to work with @PaulDatlinger and many others of the Bock lab at @CeMM_News
