Entering edit mode
Hi everyone,
I was performing some DEGs and got this message:
Warning messages:
1: In nbinomGLM(x = x, Y = YNZ, size = size, weights = weightsNZ, ... :
the line search routine failed, possibly due to insufficient numeric precision
I've realized it was something related with the negative binomial model. But how can I solve this?
Thanks.
sessionInfo()
R version 4.3.0 (2023-04-21 ucrt)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1
Matrix products: default
locale:
[1] LC_COLLATE=Portuguese_Portugal.1252 LC_CTYPE=Portuguese_Portugal.1252
[3] LC_MONETARY=Portuguese_Portugal.1252 LC_NUMERIC=C
[5] LC_TIME=Portuguese_Portugal.1252
time zone: Europe/Lisbon
tzcode source: internal
attached base packages:
[1] stats4 stats graphics grDevices utils datasets methods base
other attached packages:
[1] crayon_1.5.2 biomaRt_2.56.1 reshape2_1.4.4
[4] lubridate_1.9.2 forcats_1.0.0 stringr_1.5.0
[7] dplyr_1.1.2 purrr_1.0.1 readr_2.1.4
[10] tidyr_1.3.0 tibble_3.2.1 tidyverse_2.0.0
[13] ggplot2_3.4.2 DESeq2_1.40.1 SummarizedExperiment_1.30.2
[16] Biobase_2.60.0 MatrixGenerics_1.12.2 matrixStats_1.0.0
[19] GenomicRanges_1.52.0 GenomeInfoDb_1.36.0 IRanges_2.34.0
[22] S4Vectors_0.38.1 BiocGenerics_0.46.0
loaded via a namespace (and not attached):
[1] rstudioapi_0.14 xlsx_0.6.5 magrittr_2.0.3
[4] farver_2.1.1 rmarkdown_2.22 fs_1.6.2
[7] zlibbioc_1.46.0 vctrs_0.6.3 memoise_2.0.1
[10] RCurl_1.98-1.12 usethis_2.2.0 htmltools_0.5.5
[13] S4Arrays_1.0.4 progress_1.2.2 curl_5.0.1
[16] cellranger_1.1.0 desc_1.4.2 htmlwidgets_1.6.2
[19] plyr_1.8.8 cachem_1.0.8 mime_0.12
[22] lifecycle_1.0.3 pkgconfig_2.0.3 Matrix_1.5-4.1
[25] R6_2.5.1 fastmap_1.1.1 GenomeInfoDbData_1.2.10
[28] shiny_1.7.4 digest_0.6.31 numDeriv_2016.8-1.1
[31] colorspace_2.1-0 AnnotationDbi_1.62.1 ps_1.7.5
[34] rprojroot_2.0.3 pkgload_1.3.2 RSQLite_2.3.1
[37] filelock_1.0.2 labeling_0.4.2 fansi_1.0.4
[40] timechange_0.2.0 httr_1.4.6 compiler_4.3.0
[43] remotes_2.4.2 bit64_4.0.5 withr_2.5.0
[46] BiocParallel_1.34.2 DBI_1.1.3 pkgbuild_1.4.1
[49] MASS_7.3-60 rappdirs_0.3.3 DelayedArray_0.26.3
[52] sessioninfo_1.2.2 tools_4.3.0 httpuv_1.6.11
[55] glue_1.6.2 callr_3.7.3 promises_1.2.0.1
[58] grid_4.3.0 generics_0.1.3 gtable_0.3.3
[61] tzdb_0.4.0 hms_1.1.3 xml2_1.3.4
[64] utf8_1.2.3 XVector_0.40.0 pillar_1.9.0
[67] emdbook_1.3.12 later_1.3.1 rJava_1.0-6
[70] BiocFileCache_2.8.0 lattice_0.21-8 bit_4.0.5
[73] tidyselect_1.2.0 locfit_1.5-9.8 Biostrings_2.68.1
[76] miniUI_0.1.1.1 knitr_1.43 xfun_0.39
[79] devtools_2.4.5 stringi_1.7.12 yaml_2.3.7
[82] evaluate_0.21 codetools_0.2-19 xlsxjars_0.6.1
[85] bbmle_1.0.25 cli_3.6.1 xtable_1.8-4
[88] munsell_0.5.0 processx_3.8.1 Rcpp_1.0.10
[91] readxl_1.4.2 dbplyr_2.3.2 coda_0.19-4
[94] png_0.1-8 bdsmatrix_1.3-6 XML_3.99-0.14
[97] parallel_4.3.0 ellipsis_0.3.2 blob_1.2.4
[100] prettyunits_1.1.1 profvis_0.3.8 urlchecker_1.0.1
[103] bitops_1.0-7 mvtnorm_1.2-2 apeglm_1.22.1
[106] scales_1.2.1 rlang_1.1.1 KEGGREST_1.40.0
It's a warning, not an error. Please post a reprodicible example or at least full code. Can you make a MRE wirh
makeExampleDESeqDataset()
?Unfourtunately, I wasn't able to replicate the error with the
makeExampleDESeqDataset()
I am performing a cycle, in which I subset my dataset into organ:
Afterwards, I create another cycle to go through the comparisons:
And then I rearrange the levels to perform other comparisons of interest: