Identify cells & memory Issues

[Romeo1-1]

Hello,

Thank you for this wonderful package, it's very easy to use ! I have 2 questions:

1. I'm comparing 10 samples from 5 patients, pre and post treatment. That's the code I'm running, where I mix both pre and post samples for each patient.

pre <- CreateSeuratObject(counts = pre_data, project = "pre", min.cells = 3, min.features = 200)
post <- CreateSeuratObject(counts = post_data, project = "post", min.cells = 3, min.features = 200)

pre_post <- merge(post, y = pre, add.cell.ids = c("post", "pre"), project = 'all')

pre_post$sample <- rownames([email protected])
[email protected] <- separate([email protected], col = 'sample', 
                                 into = c("Time", 'Barcode'), 
                                 sep = '_')

pre_post_matrix = GetAssayData(object = pre_post, slot = "counts")
pre_post_matrix = as.matrix(pre_post)

results <- SCEVAN::pipelineCNA(pre_post_matrix , sample = "P1", par_cores = 20, SUBCLONES = TRUE, plotTree = TRUE)

How could I know in the results which cells are "pre" and which are "post" ? Would it be possible at least to know the proportion of pre/post cells in each cluster ? I'm interested in the lineage of those cells.

2. I encountered memory issues (R crashing) when running the pipeline with 1 patient (= 2 samples mixed together) and most importantly when running the multiSampleComparisonClonalCN with the 10 samples. I'm running the analyses on a cluster with 32 cores and 128Gb. Do you have any idea how I would be able to run the analyses while using less memories ?

Thank you a lot in advance and thanks again