neurorestore / Augur

Cell type prioritization in single-cell data
MIT License
100 stars 11 forks source link

error running single cell ATAC-seq data #27

Open jhong26 opened 1 year ago

jhong26 commented 1 year ago

Hello, I'm trying to run Augur on single-cell ATAC-seq data on human PBMCs from healthy vs disease subjects. However, I am getting the following error: "no cell type had at least 20 cells in all conditions"

library(Augur) expr=SeurObj@assays$ATAC@counts meta=SeurObj@meta.data # has columns "label" for disease vs healthy status, and "cell_type" for 29 different cell type labels. #all cell types have well over 20 cells per condition augur = calculate_auc(expr, meta) Error in calculate_auc(expr, meta) : no cell type had at least 20 cells in all conditions In addition: Warning messages: 1: In calculate_auc(expr, meta) : coercing labels to factor ... 2: In mclapply(X, FUN, ..., mc.cores = mc.cores, mc.preschedule = mc.preschedule, : scheduled cores 1, 2, 3, 4 did not deliver results, all values of the jobs will be affected Execution halted

Any input would be much appreciated. Thank you! Jason

skinnider commented 1 year ago

Hi Jason, sorry for the delay getting back to you. Tough to say without looking at your dataset but my guess would be that each process is running out of memory. You could try: (1) running on a single core to get a more informative error message (n_threads = 1) and/or (2) sending me a small subset of your dataset that reproduces the error and I can dig into it in more detail.