error running sc3_plot_de_genes

[qiaowei-vvjoe]

Hello,
When I try to run sc3_plot_de_genes(sce, k = 4), it gives the following error

Error in seq.default(min(x, na.rm = T), max(x, na.rm = T), length.out = n +  : 
  'from' must be a finite number
In addition: Warning messages:
1: In min(x, na.rm = T) : no non-missing arguments to min; returning Inf
2: In max(x, na.rm = T) : no non-missing arguments to max; returning -Inf

and also in the following function sc3_plot_markers(sce, k = 4),it gives the following error

No markers have been found, try to lower significance thresholds!

but other function sc3_plot_expression(sce, k = 4),sc3_plot_consensus(sce, k = 4)seems worked,
Could you please give me some advice about how to fix it? I am looking for the marker genes.

Thanks in advance,

[wikiselev]

looks like you probably don't have any DE or marker genes, could you check by running the interactive session? or by checking corresponding columns in the rowData?

[qiaowei-vvjoe]

@wikiselev Hi,thanks for your reply.It is my first time to issue on github so some worries about no reply.Sorry for that and I promise no more next time. Then as for the interactive Shiny session there were only 1/2/3 DE/marker genes when change some parameters and the rowData are as following:
DataFrame with 6 rows and 17 columns
sc3_gene_filter sc3_2_markers_clusts sc3_2_markers_padj sc3_2_markers_auroc

Ogfrl1 TRUE 2 1 0.594117647058824
Col9a1 FALSE NA NA NA
Inpp4a FALSE NA NA NA
Il1r1 TRUE 1 1 0.741176470588235
Nck2 FALSE NA NA NA
1500015O10Rik TRUE 1 1 0.636764705882353
sc3_3_markers_clusts sc3_3_markers_padj sc3_3_markers_auroc sc3_4_markers_clusts

Ogfrl1 3 1 0.645454545454545 2
Col9a1 NA NA NA NA
Inpp4a NA NA NA NA
Il1r1 2 0.448153711729321 0.821678321678322 4
Nck2 NA NA NA NA
1500015O10Rik 2 1 0.660839160839161 4
sc3_4_markers_padj sc3_4_markers_auroc sc3_5_markers_clusts sc3_5_markers_padj

Ogfrl1 1 0.633928571428571 2 1
Col9a1 NA NA NA NA
Inpp4a NA NA NA NA
Il1r1 1 0.795698924731183 5 1
Nck2 NA NA NA NA
1500015O10Rik 1 0.690860215053763 2 1
sc3_5_markers_auroc sc3_2_de_padj sc3_3_de_padj sc3_4_de_padj sc3_5_de_padj

Ogfrl1 0.712121212121212 1 1 1 1
Col9a1 NA NA NA NA NA
Inpp4a NA NA NA NA NA
Il1r1 0.795698924731183 1 0.348161379908233 1 1
Nck2 NA NA NA NA NA
1500015O10Rik 0.75 1 1 1 1

So my doubt is why there are so little DE/marker genes?it is because the input of my matrix?
Thank you so much in advance.

[wikiselev]

Why do you have so few genes in your matrix? Are you analysis scRNAseq or qPCR? SC3 performance for anything other than scRNAseq has not been tested at all.

[qiaowei-vvjoe]

My input matrix is 444genes37cells and it is the scRNAseq data.The gene of the matrix is choosed from the raw matrix which is 21004genes37cells.And the rowData has showed to you is just the head of rowdata which I forgot to paste the code in the previous issue.

row_data <- rowData(sce)
head(row_data[ , grep("sc3_", colnames(row_data))])
But SC3 still gives only 8 markers at most when I tested the raw matrix by changing some parameters.
I want to find more markers to find some biology significance in my project.

[wikiselev]

Well, then with such a small number of cells, most probably you just don't have any DE or marker genes. Could you share the data and your script with me on vk6@sanger.ac.uk so that I can quickly check?

[qiaowei-vvjoe]

@wikiselev Hi,I have sent email to you on Feb,25.Thank you so much in advance.

[wikiselev]

Hi, sorry for the delay, it's been two busy weeks so far and still going. I will try to look at your data next week.

[qiaowei-vvjoe]

@wikiselev OK,np.Thanks.