search

YMa-lab / CARD

GNU General Public License v3.0
80 stars 20 forks source link

sc and spatial count Matrix are correctly formatted but createCARDObject function returns error #69

Open krigia opened 3 months ago

krigia commented 3 months ago

@YingMa0107 @YMalab I have exported tha RNA counts for both scRNA-seq and spatial data as dgCMatrix.

str(scRNA_seq_counts) Formal class 'dgCMatrix' [package "Matrix"] with 6 slots ..@ i : int [1:37279034] 29 30 45 170 294 606 774 846 858 920 ... ..@ p : int [1:24475] 0 433 822 1857 2599 3661 4614 4820 5038 5707 ... ..@ Dim : int [1:2] 36601 24474 ..@ Dimnames:List of 2 .. ..$ : chr [1:36601] "MIR1302-2HG" "FAM138A" "OR4F5" "AL627309.1" ... .. ..$ : chr [1:24474] "TCa001_AAACCTGAGAAGAAGC-1" "TCa001_AAACCTGAGAAGATTC-1" "TCa001_AAACCTGAGAGTACCG-1" "TCa001_AAACCTGAGCTACCTA-1" ... ..@ x : num [1:37279034] 0.995 0.985 0.997 1.981 0.995 ... ..@ factors : list()

str(spatial_counts_dgCMatrix) Formal class 'dgCMatrix' [package "Matrix"] with 6 slots ..@ i : int [1:2804500] 1 3 8 11 14 15 18 20 22 24 ... ..@ p : int [1:2024] 0 1801 3559 5395 6542 7439 8601 9431 10073 11552 ... ..@ Dim : int [1:2] 8417 2023 ..@ Dimnames:List of 2 .. ..$ : chr [1:8417] "SAMD11" "NOC2L" "KLHL17" "HES4" ... .. ..$ : chr [1:2023] "CGGCTGAAGGTTACGC-1_1" "CACCTCTACGAGTGTG-1_1" "ATACGACAGATGGGTA-1_1" "ACTTCCTGTCGTGCGA-1_1" ... ..@ x : num [1:2804500] 1 1 1 1 1 3 1 2 1 1 ... ..@ factors : list()

However although the two dgCMatrix are properly formatted and they have overlapped genes, when I run createCARDObject function, I get the following error: **_## QC on scRNASeq dataset! ...

QC on spatially-resolved dataset! ...

Error: There are no common gene names in spatial count data and single cell RNAseq count data_**

Your advice is highly appreciated. Thank you!

YMalab commented 3 months ago

Hi @krigia,

Have you checked and printed the rownames(spatial_counts_dgCMatrix) the rownames(scRNA_seq_counts) out? Also have you checked the colSums(spatial_counts_dgCMatrix) to see if they are all very small? If they are all very small, maybe you can set the minCountGene to a smaller number.

Best, Ying

krigia commented 2 months ago

@YMalab rownames(spatial_counts_dgCMatrix) is NULL. The object has NULL values for both cells (spots) and features. Also, I do not understand what you mean to check colSums(spatial_counts_dgCMatrix) to see if they are all very small. How small? Thank you!