search

moontreegy / scseq_data_formatting

0 stars 0 forks source link

add 29545511_Mouse_Brain #26

Closed moontreegy closed 4 years ago

moontreegy commented 4 years ago

@liuyifang, refer to issue #25 Hi Yifang, This is a large dataset contains 150k+ nucleus. When I generated the dot-plot, it gives a warning “Could not find ... in the default search locations, found in RNA assay instead”, I searched the answer of the same problem in this issue: https://github.com/satijalab/seurat/issues/1198 it mentions “The "RNA" Assay stores all of original "uncorrected" data”, so I changed parameter “nfeatures=2000” to the number of all genes, but it still got the warnings. And it is weird because when I input these genes one by one, it doesn’t show the warning messages whether I change the parameter or not. Could you help me check if something wrong with the code? Thanks a lot! Yue

liuyifang commented 4 years ago

Hi Yue,

I can reproduce this problem. I think it is either because of memory or reach some limit that R can not hold. Split the gene list into smaller parts and calculate them separately, then combine the results can temporarily solve the problem.

dot_part1 <- DotPlot(object = seuratObj, features = all_genes[1:10000])
dot_part2 <- DotPlot(object = seuratObj, features = all_genes[10001:20000])
dot_part3 <- DotPlot(object = seuratObj, features = all_genes[20001:length(all_genes)])
moontreegy commented 4 years ago

refer to issue #25 I use a loop to generate the dot plots and merge them. Please review it! Thanks! @liuyifang