bit64 version 4.6.0.1 causing cryptic soma_joinid error

[sarahhp]

On write_soma with a seurat object, the following error is triggered when adding feature-level metadata:
Error in $<-.data.frame(*tmp*, "soma_joinid", value = numeric(0)) :

replacement has 0 rows, data has 100

Calls: write_soma ... tryCatch -> tryCatchList -> tryCatchOne ->

On non-default assays this is issued as a warning. Context specific; on a singularity VM no error even with the same R install. Traced the error back to # Add `soma_joinid` to `x` block in write_soma.dataframe.

Fix:
Updating bit64 package to v4.8.0 resolved the issue for me. e.g. install.packages("bit64").

Versions:

• TileDB-SOMA version: 2.3.0
• TileDB core version (libtiledbsoma): 2.30.0
• Language and language version (e.g. Python 3.9, R 4.3.2): R 4.5.0
• OS (e.g. MacOS, Ubuntu Linux): Rocky Linux 8.10 (Green Obsidian) / x86_64-pc-linux-gnu

Thanks for a great package!

[aaronwolen]

Thanks for the detailed report and for tracking down the bit64 connection!

To make sure we ship the right fix (and don't accidentally raise our bit64
floor on users who can't easily upgrade), I tried to reproduce the bug
directly. In my tests on:

• macOS arm64, R 4.5, bit64 4.6.0.1: seq(bit64::as.integer64(0L), to = N - 1L) returns proper integer64 sequences for N from 1 up to 10000.
• Rocky Linux 8.10 x86_64 (fresh EC2), R 4.5.3, bit64 4.6.0.1: same, no
  failure.

You mentioned that the bug didn't appear in a VM with the same R install, which suggests the trigger is something more specific than the bit64 version alone. If possible could you share a bit more so we can target the fix correctly?

1. Your sessionInfo() ideally captured right after the error in the same R session.
2. A reprex you used to trigger the error. If the failing assay is private, a stripped-down version (e.g., seurat_obj@assays$RNA@meta.features shape and dtypes, or a synthetic Seurat object that triggers the same error) would let us try it on the Rocky 8.10 box I already have set up.
3. The traceback from traceback() immediately after the error rather than just the error message. The exact call chain leading into seq.integer64 will tell us whether the failure is in write_soma()'s direct call or in something masking seq in your session.

Thanks again for the thorough write-up, and apologies for the extra round
of questions, I'd rather get this right than ship a guess.

[sarahhp]

Sure! Let me get that for you.

1. sessionInfo:

    	R version 4.5.0 (2025-04-11)
    	Platform: x86_64-pc-linux-gnu
    	Running under: Rocky Linux 8.10 (Green Obsidian)
    	
    	Matrix products: default
    	BLAS/LAPACK: /apps/intel-tools/.packages/2025.0.1/mkl/2025.0/lib/libmkl_intel_lp64.so.2;  LAPACK version 3.11.0
    	
    	locale:
    	 [1] LC_CTYPE=C.UTF-8       LC_NUMERIC=C           LC_TIME=C.UTF-8
    	 [4] LC_COLLATE=C.UTF-8     LC_MONETARY=C.UTF-8    LC_MESSAGES=C.UTF-8
    	 [7] LC_PAPER=C.UTF-8       LC_NAME=C              LC_ADDRESS=C
    	[10] LC_TELEPHONE=C         LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C
    	
    	time zone: :/etc/localtime
    	tzcode source: system (glibc)
    	
    	attached base packages:
    	[1] stats     graphics  grDevices utils     datasets  methods   base
    	
    	other attached packages:
    	[1] tiledbsoma_2.3.0   Seurat_5.4.0.9015  SeuratObject_5.4.0 sp_2.2-1       
    	
    	loaded via a namespace (and not attached):
    	  [1] deldir_2.0-4           pbapply_1.7-4          gridExtra_2.3
    	  [4] rlang_1.2.0            magrittr_2.0.5         RcppAnnoy_0.0.23
    	  [7] otel_0.2.0             matrixStats_1.5.0      ggridges_0.5.7
    	 [10] compiler_4.5.0         spatstat.geom_3.7-3    png_0.1-9
    	 [13] vctrs_0.7.3            reshape2_1.4.5         stringr_1.6.0
    	 [16] pkgconfig_2.0.3        fastmap_1.2.0          promises_1.5.0
    	 [19] nanotime_0.3.13        tiledb_0.34.0          bit_4.6.0
    	 [22] purrr_1.2.2            jsonlite_2.0.0         goftest_1.2-3
    	 [25] later_1.4.8            spatstat.utils_3.2-2   spdl_0.0.5
    	 [28] irlba_2.3.7            parallel_4.5.0         cluster_2.1.8.1
    	 [31] R6_2.6.1               ica_1.0-3              stringi_1.8.7
    	 [34] RColorBrewer_1.1-3     spatstat.data_3.1-9    reticulate_1.46.0
    	 [37] parallelly_1.46.1      spatstat.univar_3.1-7  lmtest_0.9-40
    	 [40] scattermore_1.2        assertthat_0.2.1       Rcpp_1.1.1-1
    	 [43] tensor_1.5.1           future.apply_1.20.2    zoo_1.8-15
    	 [46] sctransform_0.4.3      httpuv_1.6.17          Matrix_1.7-3
    	 [49] splines_4.5.0          igraph_2.2.3           tidyselect_1.2.1
    	 [52] abind_1.4-8            codetools_0.2-20       spatstat.random_3.4-5
    	 [55] miniUI_0.1.2           spatstat.explore_3.8-0 listenv_0.10.1
    	 [58] lattice_0.22-6         tibble_3.3.1           plyr_1.8.9
    	 [61] shiny_1.13.0           S7_0.2.1-1             ROCR_1.0-12
    	 [64] Rtsne_0.17             RcppCCTZ_0.2.14        future_1.70.0
    	 [67] fastDummies_1.7.5      survival_3.8-3         polyclip_1.10-7
    	 [70] fitdistrplus_1.2-6     pillar_1.11.1          KernSmooth_2.23-26
    	 [73] plotly_4.12.0          generics_0.1.4         RcppHNSW_0.6.0
    	 [76] nanoarrow_0.8.0        ggplot2_4.0.2          scales_1.4.0
    	 [79] globals_0.19.1         xtable_1.8-8           glue_1.8.0
    	 [82] lazyeval_0.2.3         tools_4.5.0            data.table_1.18.2.1
    	 [85] RSpectra_0.16-2        RANN_2.6.2             fs_2.1.0
    	 [88] dotCall64_1.2          cowplot_1.2.0          grid_4.5.0
    	 [91] tidyr_1.3.2            nlme_3.1-168           patchwork_1.3.2
    	 [94] RcppSpdlog_0.0.28      cli_3.6.6              spatstat.sparse_3.1-0
    	 [97] spam_2.11-3            viridisLite_0.4.3      arrow_23.0.1.2
    	[100] dplyr_1.2.1            uwot_0.2.4             gtable_0.3.6
    	[103] digest_0.6.39          progressr_0.19.0       ggrepel_0.9.8
    	[106] htmlwidgets_1.6.4      farver_2.1.2           htmltools_0.5.9
    	[109] lifecycle_1.0.5        httr_1.4.8             mime_0.13
   

[112] bit64_4.6.0-1 MASS_7.3-65

2. My test object can be produced with:

rna_counts <- matrix(
rpois(2000, 5),
nrow = 100,
dimnames = list(paste0("gene", 1:100), paste0("cell", 1:20))
)
so <- suppressWarnings(CreateSeuratObject(counts = rna_counts))
so <- NormalizeData(so, verbose = FALSE)
so <- FindVariableFeatures(so, verbose = FALSE)
so <- ScaleData(so, verbose = FALSE)

PCA and UMAP dim reductions

so <- RunPCA(so, npcs = 5, verbose = FALSE)
so[["umap"]] = so[["pca"]]
so@reductions$umap

A second assay (e.g. simulated ADT protein counts)

adt_counts <- matrix(
rpois(200, 10),
nrow = 10,
dimnames = list(paste0("protein", 1:10), paste0("cell", 1:20))
)
so[["ADT"]] <- CreateAssayObject(counts = adt_counts)

saveRDS(so, here::here("R/tests/rich_seurat.rds"))
so

3. Traceback:

[sarahhp]

Interestingly enough the Rstudio VM that's working with the older bt64 version has an almost identical sessionInfo: R version 4.5.0 (2025-04-11)
Platform: x86_64-pc-linux-gnu
Running under: Rocky Linux 8.10 (Green Obsidian)

		Matrix products: default
		BLAS/LAPACK: /apps/intel-tools/.packages/2025.0.1/mkl/2025.0/lib/libmkl_intel_lp64.so.2;  LAPACK version 3.11.0
		
		locale:
		 [1] LC_CTYPE=C.UTF-8       LC_NUMERIC=C           LC_TIME=C.UTF-8
		 [4] LC_COLLATE=C.UTF-8     LC_MONETARY=C.UTF-8    LC_MESSAGES=C.UTF-8
		 [7] LC_PAPER=C.UTF-8       LC_NAME=C              LC_ADDRESS=C
		[10] LC_TELEPHONE=C         LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C
		
		time zone: :/etc/localtime
		tzcode source: system (glibc)
		
		attached base packages:
		[1] stats     graphics  grDevices utils     datasets  methods   base
		
		other attached packages:
		[1] tiledbsoma_2.3.0   Seurat_5.4.0.9015  SeuratObject_5.4.0 sp_2.2-1       
		
		loaded via a namespace (and not attached):
		  [1] deldir_2.0-4           pbapply_1.7-4          gridExtra_2.3
		  [4] rlang_1.2.0            magrittr_2.0.5         RcppAnnoy_0.0.23
		  [7] otel_0.2.0             matrixStats_1.5.0      ggridges_0.5.7
		 [10] compiler_4.5.0         spatstat.geom_3.7-3    png_0.1-9
		 [13] vctrs_0.7.3            reshape2_1.4.5         stringr_1.6.0
		 [16] pkgconfig_2.0.3        fastmap_1.2.0          promises_1.5.0
		 [19] nanotime_0.3.13        tiledb_0.34.0          bit_4.6.0
		 [22] purrr_1.2.2            jsonlite_2.0.0         goftest_1.2-3
		 [25] later_1.4.8            spatstat.utils_3.2-2   spdl_0.0.5
		 [28] irlba_2.3.7            parallel_4.5.0         cluster_2.1.8.1
		 [31] R6_2.6.1               ica_1.0-3              stringi_1.8.7
		 [34] RColorBrewer_1.1-3     spatstat.data_3.1-9    reticulate_1.46.0
		 [37] parallelly_1.46.1      spatstat.univar_3.1-7  lmtest_0.9-40
		 [40] scattermore_1.2        assertthat_0.2.1       Rcpp_1.1.1-1
		 [43] tensor_1.5.1           future.apply_1.20.2    zoo_1.8-15
		 [46] sctransform_0.4.3      httpuv_1.6.17          Matrix_1.7-3
		 [49] splines_4.5.0          igraph_2.2.3           tidyselect_1.2.1
		 [52] abind_1.4-8            codetools_0.2-20       spatstat.random_3.4-5
		 [55] miniUI_0.1.2           spatstat.explore_3.8-0 listenv_0.10.1
		 [58] lattice_0.22-6         tibble_3.3.1           plyr_1.8.9
		 [61] shiny_1.13.0           S7_0.2.1-1             ROCR_1.0-12
		 [64] Rtsne_0.17             RcppCCTZ_0.2.14        future_1.70.0
		 [67] fastDummies_1.7.5      survival_3.8-3         polyclip_1.10-7
		 [70] fitdistrplus_1.2-6     pillar_1.11.1          KernSmooth_2.23-26
		 [73] plotly_4.12.0          generics_0.1.4         RcppHNSW_0.6.0
		 [76] nanoarrow_0.8.0        ggplot2_4.0.2          scales_1.4.0
		 [79] globals_0.19.1         xtable_1.8-8           glue_1.8.0
		 [82] lazyeval_0.2.3         tools_4.5.0            data.table_1.18.2.1
		 [85] RSpectra_0.16-2        RANN_2.6.2             fs_2.1.0
		 [88] dotCall64_1.2          cowplot_1.2.0          grid_4.5.0
		 [91] tidyr_1.3.2            nlme_3.1-168           patchwork_1.3.2
		 [94] RcppSpdlog_0.0.28      cli_3.6.6              spatstat.sparse_3.1-0
		 [97] spam_2.11-3            viridisLite_0.4.3      arrow_23.0.1.2
		[100] dplyr_1.2.1            uwot_0.2.4             gtable_0.3.6
		[103] digest_0.6.39          progressr_0.19.0       ggrepel_0.9.8
		[106] htmlwidgets_1.6.4      farver_2.1.2           htmltools_0.5.9
		[109] lifecycle_1.0.5        httr_1.4.8             mime_0.13

[112] bit64_4.6.0-1 MASS_7.3-65