FindIntegrationAnchors returs 0 anchor after filtering
0
0
Entering edit mode
14 months ago
mbk0asis ▴ 640

Hi, all~

I have two seurat objects that I want to integrate.

> seu
An object of class Seurat 
22663 features across 66 samples within 1 assay 
Active assay: RNA (22663 features, 2000 variable features)
 2 dimensional reductions calculated: pca, umap

> mipsc
An object of class Seurat 
20391 features across 33966 samples within 1 assay 
Active assay: RNA (20391 features, 2000 variable features)
 2 dimensional reductions calculated: pca, umap

A problem arose when I ran 'FindIntegrationAnchors'.

I ran with the default setting it throwed an error complaining about 'cell number' and 'k.filter'.

So I lower the parameters and passed this step without error.

> anchors <- FindIntegrationAnchors(object.list = seu.list, 
                                   anchor.features = features,
                                   dims = 1:5, k.filter=1, k.score=1)
Scaling features for provided objects
  |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=03s  
Finding all pairwise anchors
  |                                                  | 0 % ~calculating  Running CCA
Merging objects
Finding neighborhoods
Finding anchors
    Found 330 anchors
Filtering anchors
    Retained 0 anchors
  |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=25s  

But soon, I found the retained anchor was 0, and of course the integration failed with another error.

> combined <- IntegrateData(anchorset = anchors, k.weight = 1)
Error in hclust(d = dist.mat) : 
  NA/NaN/Inf in foreign function call (arg 10)

Is this something to do with sample size difference between two data because it worked when I integrated using a much smaller dataset with 16 samples.

How can I solve this problem?

Thank you!

seurat dataintegration FindIntegrationAnchors • 1.1k views
ADD COMMENT
0
Entering edit mode

Do they have the same gene names between the two Seurat objects?

ADD REPLY
0
Entering edit mode

Hi~ ATpoint!

I checked the gene names and found 16,174 genes successfully matched.

Thank you!

ADD REPLY

Login before adding your answer.

Traffic: 1676 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