Question: sc-RNA data analysis using scater
gravatar for hrishi27n
16 months ago by
hrishi27n0 wrote:

Hello All,

I am trying to analyze data for a single cell RNA sequencing experiment, for QC and normalization I am considering using the scater package. There are a few things I would like to know before starting analyzing this dataset. All your help and suggestions are much appreciated. This is my first attempt to analyze sc-RNA, I apologize in advance if my questions are confusing. Questions: 1) The sequencing lab is using an unpublished protocol, they have provided read counts and spike-ins file separately. Do I need to combine these two files? I am considering this, for the "feature_controls" option for calculateQCMetrics method.

2) After doing the initial QC, I see that total counts for all my wells(cells) is almost >50k+ and the number of genes detected is above 10k. I am removing genes that have 0 expression, I am also filtering genes with very low average nonzero expression across all cells(using a mean of counts across all cells). Do I need to do any other filtering for both cells and genes?

ADD COMMENTlink modified 16 months ago by Aaron Lun19k • written 16 months ago by hrishi27n0
gravatar for Aaron Lun
16 months ago by
Aaron Lun19k
Cambridge, United Kingdom
Aaron Lun19k wrote:

For your first question; does the "read counts" file already contain counts for the spike-in transcripts? If yes, then that's all you need to make a SCESet object in scater; just supply the count matrix as countData= in the constructor. Otherwise, you'll first have to rbind the matrix of gene counts with that of the spike-in counts. Note that you don't need to know the concentrations of the spike-ins to use most scater functions.

As for your second question, have a look at

ADD COMMENTlink modified 16 months ago • written 16 months ago by Aaron Lun19k


Thanks for the reply. My "read count" file does't include the ERCC's. A grep for "^ERCC-" doesn't really give anything, however the ERCC's are provided in a separate file. I guess I might have to do a rbind on the gene count file to include the spike-in data. 

ADD REPLYlink written 16 months ago by hrishi27n0
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 120 users visited in the last hour