about DropletUtils and scRNA-seq
1
0
Entering edit mode
Bogdan ▴ 660
@bogdan-2367
Last seen 1 day ago
Palo Alto, CA, USA

Dear Aaron, and all, happy and healthy new year !

I was following the tutorial presented at :

https://master.bioconductor.org/packages/release/workflows/vignettes/simpleSingleCell/inst/doc/work-3-tenx.html

may I ask, what function do you use in order to remove from a SCE object the barcodes with low total UMI counts

(i.e the empty droplets) ? thanks a lot !

 

-- bogdan

 

dropseq dropletutils droplet • 620 views
ADD COMMENT
0
Entering edit mode

... just found the function in the tutorial together with the explanations of the Monte Carlo procedure and the choice of FDR < 0.01 (sorry, 've overlooked it ..)

sce <- sce[,which(e.out$FDR <= 0.01)]
ADD REPLY
0
Entering edit mode
Aaron Lun ★ 28k
@alun
Last seen 3 hours ago
The city by the bay

Looks like you solved your own problem. Note that 1% is just a threshold I've used; if you don't like that level of error, you can (and should) set it lower. For example, the BioC-devel versions of the workflows use 0.1%.

ADD COMMENT
0
Entering edit mode

Thank you Aaron. If I may add a question about scRNA-seq analysis (of 10X_Genomics datasets) :

we have 3 experimental conditions to compare (i.e. 3 scRNA-seq datasets : control, treatment1, treatment2).

when I analyze each dataset, i could apply the pipelines that you comprehensively described at :

https://master.bioconductor.org/packages/release/workflows/vignettes/simpleSingleCell/inst/doc/work-3-tenx.html

however, shall i compare the specific cell subpopulations in CONTROL vs TREATMENT1 (or TREATMENT2),

which workflow shall I follow (in order to be able to find differentially expressed genes in specific cells) ?

thanks a lot, happy weekend !

ADD REPLY
1
Entering edit mode

Follow the instructions in the batch workflow to merge all data sets together; cluster cells into subpopulations; and then perform differential expression between conditions within each cluster. This will allow you to identify, e.g., cell-type-specific changes in expression for each cluster. You can also test for differential abundance within each cluster to identify changes in population composition between conditions, a la mass cytometry. I will write up a more detailed workflow on how to do this next month. 

ADD REPLY
0
Entering edit mode

Dear Aaron, thank you very much.

Having a workflow that will give a more detailed example on how to perform differential expression :

1) between conditions within each cluster, and 2) between conditions of different clusters,

would be very very helpful. For our scRNA data analysis, we would like to use at least 2 workflows  :

-- an workflow based on SingleCellExperiment packages that you've written, including "scatter", and "scran"

-- an workflow based on Seurat, at outlined here : https://satijalab.org/seurat/immune_alignment.html

 

ADD REPLY

Login before adding your answer.

Traffic: 173 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6