Question: Merging Seurat objects together and gene alignment
gravatar for amjass
16 months ago by
amjass20 wrote:

Hi all, I was wondering if somebody could help with a question i have regarding the seurat pipeline i am running for single cell RNA-seq analysis.

I was just wondering how merging two or more Seurat object handles the objects when gene lists are of different lengths (i,.e one of the object with a count matrix of 18,000 and the other of 20,000 for example. I have a couple of questions just for my own sanity to be sure that merging the objects is working correctly:

Do the genes have to be in the same order for both seurat objects before merging ? (presumably not as the merge function should match values) and therefore counts for each gene in different object should be correct for all samples?

For some seurat objects that are annotated with gene names and not ensembl IDs, some gene names overlap and so i simply make these unique when reading them in, so there will be instances where there is gene x and x.1. How does the merge function handle these genes as well as genes that are present in one data set but not the other?

Are the counts that exist for one data set added to the bottom of the list and then for the other seurat object all assigned 0's for those counts?

Does merging two seurat objects which represent different experiments take care of the above? I have not encountered this in the vignette or other documentation... In the vignette, it seems as as simple as simply merging objects together and there is no mention of different gene lengths between datasets so i am guessing this is taken care of?

I dont want to be faced with the situation where genes are misaligned when merging together and therefore counts are no longer what they should be!!

does all.genes <- rownames(Seuratmerged) still keep the right genes and counts for each sample correctly?

I dont have more extensive code as this might just be answered with a yes that is taken care of (in the perfect world haha).. but if you need see code to specifically answer the question then I am more than happy to provide it!!

Any help would be much appreciated! thank you!

rna-seq seurat single-cell • 1.4k views
ADD COMMENTlink modified 16 months ago • written 16 months ago by amjass20

It sounds like you already considered many of the potential issues that may arise. Have you tried merging and checking what happens?

ADD REPLYlink written 16 months ago by igor11k
gravatar for amjass
16 months ago by
amjass20 wrote:

Yes thankyou! I can confirm that merge order does not matter and that gene present in one data set but not the other are simply appended.. and for the dataset in which they are missing, a value of 0 is assigned for those genes!

I posted this question also on Seurat/github and this was confirmed:)

ADD COMMENTlink written 16 months ago by amjass20
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1599 users visited in the last hour