doHMRF

[sfpacman]

I ran HMRF on two Visium HD samples . The first sample completed successfully, but the second one failed. The second one is about twice the size of the first one (74708 vs 33510). However, both runs did not appear to use much memory or CPU, so I am unsure if the failure is related to the larger sample size

coexpr_dt <- data.table::data.table(
  genes = names(spat_cor_netw_DT$cor_clusters$spat_netw_clus),
  cluster = spat_cor_netw_DT$cor_clusters$spat_netw_clus)

data.table::setorder(coexpr_dt, cluster)

top50_coexpr_dt <- coexpr_dt[, head(.SD, 30) , by = cluster]

spatial_genes <- top50_coexpr_dt$genes

out_folder="HMRF/control"
HMRF_spatial_genes <- doHMRF(gobject = giotto_obj,
                             expression_values = "scaled",
                             spatial_genes = spatial_genes, 
                             k = 20,
                             spatial_network_name = "Delaunay_network",
                             betas = c(0, 10, 5),
                             output_folder =out_folder )

Error message

[1] "/home/usr1/miniforge3/envs/giotto_env/bin/python /home/usr1/Analysis/kidney_rejection_spatial/seurat_analysis/renv/library/linux-ubuntu-jammy/R-4.4/x86_64-pc-linux-gnu/Giotto/python/reader2.py -l \"HMRF/control/spatial_cell_locations.txt\" -g \"HMRF/control/spatial_genes.txt\" -n \"HMRF/control/spatial_network.txt\" -e \"HMRF/control/expression_matrix.txt\" -o \"HMRF/control/result.spatial.zscore\" -a test -k 20 -b 0 10 5 -t 1e-10 -z none -s 100 -i 100"
openjdk version "1.8.0_412"
OpenJDK Runtime Environment (Zulu 8.78.0.19-CA-linux64) (build 1.8.0_412-b08)
OpenJDK 64-Bit Server VM (Zulu 8.78.0.19-CA-linux64) (build 25.412-b08, mixed mode)
Segmentation fault (core dumped)

> packageVersion("Giotto")
[1] ‘4.2.2’

[sfpacman]

It seems like the segmentation fault has to do with the size of the data. I was able to ran HMRF after downsampling .