SpiecEasi 1.99.5
A common issue that comes up when running spiec.easi is coming up with an empty network after running StARS.
For example:
library(SpiecEasi)
data(amgut1.filt)
pargs <- list(seed=10010)
se3 <- spiec.easi(amgut1.filt, method='mb', lambda.min.ratio=5e-1, nlambda=10, pulsar.params=pargs)getOptInd(se3)
# [1] 1
sum(getRefit(se3))/2
# [1] 139As the warning indicates, the network stability could not be determined from the lambda path. Looking at the stability along the lambda path, se$select$stars$summary, we can see that the maximum value of the StARS summary statistic never crosses the default threshold (0.05).
This problem we can fix by lowering lambda.min.ratio to explore denser networks:
se4 <- spiec.easi(amgut1.filt, method='mb', lambda.min.ratio=1e-1, nlambda=10, pulsar.params=pargs)We have now fit a network, but since we have only a rough, discrete sampling of networks along the lambda path, we should check how far we are from the target stability threshold (0.05):
getStability(se4)
# [1] 0.0003237095
sum(getRefit(se4))/2
# [1] 158To get closer to the mark, we should bump up nlambda to more finely sample of the lambda path, which gives a denser network:
se5 <- spiec.easi(amgut1.filt, method='mb', lambda.min.ratio=1e-1, nlambda=100, pulsar.params=pargs)getStability(se5)
# [1] 0.0003237095
sum(getRefit(se5))/2
# [1] 210Problem: After running spiec.easi, you get an empty network (no edges).
Solutions:
- Lower lambda.min.ratio to explore denser networks
- Increase nlambda for finer sampling of the lambda path
- Check if your data has sufficient signal-to-noise ratio
- Try different methods (‘mb’ vs ‘glasso’)
Problem: The inferred network has too many edges.
Solutions:
- Increase lambda.min.ratio to explore sparser networks
- Adjust the StARS threshold in pulsar.params
- Use cross-validation instead of StARS
Problem: The analysis takes too long or runs out of memory.
Solutions:
- Use parallel processing with ncores parameter (Unix-like systems only)
- Use B-StARS method for large datasets
- Reduce rep.num in pulsar.params
- Use batch mode for HPC clusters
Problem: Error “‘mc.cores’ > 1 is not supported on Windows”
Solutions:
- Use ncores=1 for serial processing on Windows
- Use snow cluster for parallel processing on Windows:
library(parallel)
cl <- makeCluster(4, type = "SOCK")
pargs.windows <- list(rep.num=50, seed=10010, cluster=cl)
se.windows <- spiec.easi(data, method='mb', pulsar.params=pargs.windows)
stopCluster(cl)Problem: The algorithm doesn’t converge or gives warnings.
Solutions: - Check data preprocessing and normalization - Ensure data doesn’t have constant columns - Try different starting values - Check for missing or infinite values
Problem: R runs out of memory during analysis.
Solutions: - Use sparse matrices where possible - Reduce dataset size by filtering rare taxa - Use batch processing for large datasets - Increase system memory if available
mc.cores > 1) is not supportedncores=1 for serial processingmc.coresncores parameter directlySpiecEasi provides several functions to help diagnose issues:
# Check stability along lambda path
getStability(se)
# Get optimal lambda index
getOptInd(se)
# Get summary statistics
se$select$stars$summary
# Check network density
sum(getRefit(se))/2
# Visualize stability curve
plot(se$select$stars$summary)
# Check platform information
.Platform$OS.typelambda.min.ratio = 1e-2nlambda = 20-50rep.num = 20-50lambda.min.ratio = 1e-3nlambda = 50-100rep.num = 50-100lambda.min.ratio = 1e-4nlambda = 100+rep.num = 100+ncores=1 for serial processingSession info:
sessionInfo()
# R version 4.6.0 alpha (2026-04-05 r89794)
# Platform: x86_64-pc-linux-gnu
# Running under: Ubuntu 24.04.4 LTS
#
# Matrix products: default
# BLAS: /home/biocbuild/bbs-3.23-bioc/R/lib/libRblas.so
# LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.12.0 LAPACK version 3.12.0
#
# locale:
# [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
# [3] LC_TIME=en_GB LC_COLLATE=C
# [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
# [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
# [9] LC_ADDRESS=C LC_TELEPHONE=C
# [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
#
# time zone: America/New_York
# tzcode source: system (glibc)
#
# attached base packages:
# [1] stats graphics grDevices utils datasets methods base
#
# other attached packages:
# [1] phyloseq_1.55.2 igraph_2.2.3 Matrix_1.7-5 SpiecEasi_1.99.5
# [5] BiocStyle_2.39.0
#
# loaded via a namespace (and not attached):
# [1] ade4_1.7-24 tidyselect_1.2.1
# [3] dplyr_1.2.1 farver_2.1.2
# [5] Biostrings_2.79.5 S7_0.2.1
# [7] fastmap_1.2.0 digest_0.6.39
# [9] lifecycle_1.0.5 cluster_2.1.8.2
# [11] survival_3.8-6 magrittr_2.0.5
# [13] compiler_4.6.0 rlang_1.2.0
# [15] sass_0.4.10 tools_4.6.0
# [17] yaml_2.3.12 data.table_1.18.2.1
# [19] knitr_1.51 labeling_0.4.3
# [21] S4Arrays_1.11.1 DelayedArray_0.37.1
# [23] plyr_1.8.9 RColorBrewer_1.1-3
# [25] abind_1.4-8 withr_3.0.2
# [27] BiocGenerics_0.57.0 grid_4.6.0
# [29] stats4_4.6.0 multtest_2.67.0
# [31] biomformat_1.39.16 ggplot2_4.0.2
# [33] scales_1.4.0 iterators_1.0.14
# [35] MASS_7.3-65 dichromat_2.0-0.1
# [37] tinytex_0.59 SummarizedExperiment_1.41.1
# [39] cli_3.6.5 rmarkdown_2.31
# [41] vegan_2.7-3 crayon_1.5.3
# [43] generics_0.1.4 otel_0.2.0
# [45] pulsar_0.3.13 reshape2_1.4.5
# [47] ape_5.8-1 cachem_1.1.0
# [49] stringr_1.6.0 splines_4.6.0
# [51] parallel_4.6.0 BiocManager_1.30.27
# [53] XVector_0.51.0 matrixStats_1.5.0
# [55] vctrs_0.7.2 glmnet_4.1-10
# [57] jsonlite_2.0.0 VGAM_1.1-14
# [59] bookdown_0.46 IRanges_2.45.0
# [61] S4Vectors_0.49.1 magick_2.9.1
# [63] foreach_1.5.2 jquerylib_0.1.4
# [65] glue_1.8.0 codetools_0.2-20
# [67] stringi_1.8.7 shape_1.4.6.1
# [69] gtable_0.3.6 GenomicRanges_1.63.2
# [71] tibble_3.3.1 pillar_1.11.1
# [73] htmltools_0.5.9 Seqinfo_1.1.0
# [75] huge_1.5.1 R6_2.6.1
# [77] evaluate_1.0.5 lattice_0.22-9
# [79] Biobase_2.71.0 bslib_0.10.0
# [81] Rcpp_1.1.1 permute_0.9-10
# [83] SparseArray_1.11.13 nlme_3.1-169
# [85] mgcv_1.9-4 xfun_0.57
# [87] MatrixGenerics_1.23.0 pkgconfig_2.0.3