0
$\begingroup$

Seurat V2 had a option to find clustering information saved in object: PrintFindClustersParams(object = pbmc). How can I get the same clustering parameters from objects in Seurat3? Do I need to manually save clustering parameters or they are saved automatically?

Update: the command line object@commands works perfectly for ...$FindClusters. However, if I would like to find information about dimensions used there is an error:

...$FindNeighbors.SCT.pca Error in [[.Seurat(x, i, drop = TRUE) : Cannot find 'FindNeighbors.SCT.pca' in this Seurat object
$\endgroup$
1
$\begingroup$

I am not sure of the output of PrintFindClustersParams, as I haven't use it before, but try with object@commands. It returns a list containing every command and parameters used.

For example

...
$FindNeighbors.SCT.pca
Command: FindNeighbors(seurObj, dims = 1:20, verbose = vrb)
Time: 2019-12-11 12:41:12
reduction : pca 
dims : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 
assay : SCT 
k.param : 20 
compute.SNN : TRUE 
prune.SNN : 0.06666667 
nn.method : rann 
annoy.metric : euclidean 
nn.eps : 0 
verbose : FALSE 
force.recalc : FALSE 
do.plot : FALSE 
graph.name : SCT_nn SCT_snn 

$FindClusters
Command: FindClusters(seurObj, verbose = vrb)
Time: 2019-12-11 12:41:17
graph.name : SCT_snn 
modularity.fxn : 1 
resolution : 0.8 
algorithm : 1 
n.start : 10 
n.iter : 10 
random.seed : 0 
group.singletons : TRUE 
verbose : FALSE 
...

Re: update

I think it depends on how you built the object. FindNeighbors.SCT.pca should be result of running sctransform normalization. If you use the standard way, you may have other names (e.g.FindNeighbors.pca).

$\endgroup$
0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.