Calculating differential expression genes between two sample sequenced using 10x

Question

[singcell]

I have treatment and control samples single cell sequencing performed using 10x method. I wanted to know how to calculate the differential expression of genes between two groups. I am trying to use this tutorial from seurat but it does not say how to import two different sample dataset into seurat. Can someone recommend easier way to perform this analysis in single cell experiments.

Answer 1

[Kristoffer Vitting-Seerup]

Taking the time out to follow the Seurat tutorial is worth it. Single cell data have so many biases and problems you will need to have under control for your differential expression analysis to be trustworthy.

With regards to your data I would recommend using Seurat3's data integration to create a joint dataset - a nice tutorial equivalent to your data can be found here. Afterwards you can follow the tutorial you refere to above. An alternative would be CONOS - you can find that tutorial here. Furthermore if you have multiple replicates for your data you could consider muscat.

[singcell]

Kristoffer, Thank you for your reply. I am trying to use Seurat3 data integration to create joint dataset using the link you provided. However, they are using a test dataset 'ifnb' distributed through their SeuratData package and don't give much details on how to generate this dataset from raw 10x output. I have 'cellranger' count output files such as 'bascodes.tsv', 'features.tsv' and 'matrix.mtx' but I am not sure how to process these cellranger output files to feed to this Seurat3 pipeline. Also how to process and merge multiple files from control and treated samples.

[Kristoffer Vitting-Seerup]

You can use Seurat's Read10X() function to get the 10X data into R. It is true you might want to extend the processing part in the link above with the standard processing and QC from this vignette. If that was not what you meant could you be a bit more specific about what your problem is?