search

rezakj / iCellR

Single (i) Cell R package (iCellR) is an interactive R package to work with high-throughput single cell sequencing technologies (i.e scRNA-seq, scVDJ-seq, scATAC-seq, CITE-Seq and Spatial Transcriptomics (ST)).
121 stars 19 forks source link

an error message - running the run.knetl #37

Open MavroLucic opened 10 months ago

MavroLucic commented 10 months ago

Hi guys

I want to test the KNetL map for some geochemical purposes (sparse geochemical data 5600 rows and 19 columns). However, I have some problems, when I try to run the run.knetl, I get an error message.

This is my code:

library(iCellR) my.obj <- make.obj(data)

Filtering the data my.obj <- cell.filter(my.obj) Normalizing the data my.obj <- norm.data(my.obj, norm.method = "ranked.glsf")

my.obj <- gene.stats(my.obj, which.data = "main.data")

Run PCA my.obj <- run.pca(my.obj, method = "base.mean.rank", data.type = "main") opt.pcs.plot(my.obj)

Run DR my.obj <- run.pc.tsne(my.obj, dims = 1:10, perplexity = 6) my.obj <- run.umap(my.obj, dims = 1:10) my.obj <- run.knetl(my.obj, dims = 1:10, zoom = 300, dim.redux = "umap")

I got an error here

Getting PCA data Calculating euclidean distance ... Finding 300 neighboring cells per cell ... Generating graph from root to neighboring cells ... Generating 2D Layouts ... Error in dimnames(x) <- dn : length of 'dimnames' [1] not equal to array extent In addition: Warning messages: 1: In graph.data.frame(data, directed = FALSE) : In d'NA' elements were replaced with string "NA" 2: In get(layout.2d)(g) : Non-positive edge weight found, ignoring all weights during graph layout.

Can somebody help me?

Thank you in advance. Kind regards

rezakj commented 10 months ago

You have 19 columns. If it was single cell data it would be 19 cells. I don't recomend running KNetL map on small data. Generally it dose better when you have at least a couple of thousands of cells/columns. When you have less than 1000 cells, UMAP works better but in your case with only 19 columns, I don't think iCellR would be right for you.

Reza