startVsEndTest yields results for non existing pseudotimeValues
sam-israel opened this issue · 1 comments
sam-israel commented
I have run fitGAM on slingshot results, within one lineage, with 3 knots.
Note: this specific choice of knots is due to the fact that there are precisely 3 cell states in which I am interested - in the beginning, end, and in the middle (according to the slingshot DPT).
The DPT in the slingshot trajectory ranges from 0 to 0.1.
However, running startVsEndTest(pseudotimeValues=c(0.1,0.3)) does yield results.
Col1slingRes_filt_GAM@metadata$tradeSeq$knots
0% 50% 100%
0.0000000 0.0608644 0.1024949
(res <- Col1slingRes_filt_GAM %>% startVsEndTest(pseudotimeValues = c(0.1, 0.3))%>% mutate(padj = p.adjust(pvalue, method="BH")) %>% filter(padj<0.05))
waldStat | df | pvalue| logFClineage1 |
ENSG00000176022 | 21.121796 | 1 | 4.309978e-06 | 21.4182296 |
ENSG00000175756 | 12.047226 | 1 | 5.186952e-04 | -8.2553983 |
ENSG00000242485| 15.522301 | 1 | 8.153764e-05 | -8.8633407 |
ENSG00000160075 | 12.472358 | 1 | 4.130184e-04 | 8.4384998 |
...
res %>% nrow()
[1] 2611
...
Session info
R version 4.3.0 (2023-04-21)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 20.04.6 LTS
Matrix products: default
BLAS: /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.9.0
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.9.0
locale:
[1] LC_CTYPE=en_IL.UTF-8 LC_NUMERIC=C LC_TIME=en_IL.UTF-8
[4] LC_COLLATE=en_IL.UTF-8 LC_MONETARY=en_IL.UTF-8 LC_MESSAGES=en_IL.UTF-8
[7] LC_PAPER=en_IL.UTF-8 LC_NAME=C LC_ADDRESS=C
[10] LC_TELEPHONE=C LC_MEASUREMENT=en_IL.UTF-8 LC_IDENTIFICATION=C
time zone: Asia/Jerusalem
tzcode source: system (glibc)
attached base packages:
[1] grid stats4 stats graphics grDevices utils datasets methods base
other attached packages:
[1] sva_3.48.0 BiocParallel_1.34.2 genefilter_1.82.1
[4] mgcv_1.8-42 nlme_3.1-162 RColorBrewer_1.1-3
[7] pheatmap_1.0.12 ggrepel_0.9.3 VennDiagram_1.7.3
[10] futile.logger_1.4.3 traviz_1.6.0 enrichR_3.2
[13] tradeSeq_1.13.01 slingshot_2.8.0 TrajectoryUtils_1.8.0
[16] SingleCellExperiment_1.22.0 SummarizedExperiment_1.30.2 Biobase_2.60.0
[19] GenomicRanges_1.52.0 GenomeInfoDb_1.36.0 IRanges_2.34.0
[22] S4Vectors_0.38.1 BiocGenerics_0.46.0 MatrixGenerics_1.12.2
[25] matrixStats_1.0.0 princurve_2.1.6 glue_1.6.2
[28] destiny_3.14.0 ggplot2_3.4.2 dplyr_1.1.2
[31] SeuratObject_4.1.3 Seurat_4.3.0
loaded via a namespace (and not attached):
[1] spatstat.sparse_3.0-1 bitops_1.0-7 httr_1.4.6 tools_4.3.0
[5] sctransform_0.3.5 utf8_1.2.3 R6_2.5.1 lazyeval_0.2.2
[9] uwot_0.1.14 withr_2.5.0 sp_1.6-1 gridExtra_2.3
[13] progressr_0.13.0 cli_3.6.1 formatR_1.14 spatstat.explore_3.2-1
[17] sass_0.4.6 robustbase_0.99-0 spatstat.data_3.0-1 proxy_0.4-27
[21] ggridges_0.5.4 pbapply_1.7-0 parallelly_1.36.0 WriteXLS_6.4.0
[25] limma_3.56.2 TTR_0.24.3 rstudioapi_0.14 RSQLite_2.3.1
[29] generics_0.1.3 ica_1.0-3 spatstat.random_3.1-5 car_3.1-2
[33] Matrix_1.5-4.1 fansi_1.0.4 abind_1.4-5 lifecycle_1.0.3
[37] scatterplot3d_0.3-44 yaml_2.3.7 edgeR_3.42.4 carData_3.0-5
[41] Rtsne_0.16 blob_1.2.4 promises_1.2.0.1 crayon_1.5.2
[45] miniUI_0.1.1.1 lattice_0.21-8 cowplot_1.1.1 annotate_1.78.0
[49] KEGGREST_1.40.0 pillar_1.9.0 knitr_1.43 rjson_0.2.21
[53] boot_1.3-28.1 future.apply_1.11.0 codetools_0.2-19 leiden_0.4.3
[57] pcaMethods_1.92.0 data.table_1.14.8 vcd_1.4-11 vctrs_0.6.3
[61] png_0.1-8 gtable_0.3.3 cachem_1.0.8 xfun_0.39
[65] S4Arrays_1.0.4 mime_0.12 RcppEigen_0.3.3.9.3 rsconnect_0.8.29
[69] survival_3.5-5 rgl_1.2.1 ellipsis_0.3.2 fitdistrplus_1.1-11
[73] ROCR_1.0-11 xts_0.13.1 bit64_4.0.5 RcppAnnoy_0.0.20
[77] bslib_0.5.0 irlba_2.3.5.1 KernSmooth_2.23-21 colorspace_2.1-0
[81] DBI_1.1.3 nnet_7.3-19 tidyselect_1.2.0 smoother_1.1
[85] bit_4.0.5 compiler_4.3.0 curl_5.0.1 DelayedArray_0.26.3
[89] plotly_4.10.2 scales_1.2.1 DEoptimR_1.0-14 lmtest_0.9-40
[93] hexbin_1.28.3 stringr_1.5.0 digest_0.6.31 goftest_1.2-3
[97] spatstat.utils_3.0-3 rmarkdown_2.22 XVector_0.40.0 htmltools_0.5.5
[101] pkgconfig_2.0.3 base64enc_0.1-3 fastmap_1.1.1 rlang_1.1.1
[105] htmlwidgets_1.6.2 ggthemes_4.2.4 shiny_1.7.4 jquerylib_0.1.4
[109] zoo_1.8-12 jsonlite_1.8.5 RCurl_1.98-1.12 magrittr_2.0.3
[113] GenomeInfoDbData_1.2.10 patchwork_1.1.2 munsell_0.5.0 Rcpp_1.0.10
[117] viridis_0.6.3 reticulate_1.30 stringi_1.7.12 zlibbioc_1.46.0
[121] MASS_7.3-60 plyr_1.8.8 parallel_4.3.0 listenv_0.9.0
[125] deldir_1.0-9 Biostrings_2.68.1 splines_4.3.0 tensor_1.5
[129] locfit_1.5-9.8 igraph_1.5.0 ranger_0.15.1 spatstat.geom_3.2-1
[133] RcppHNSW_0.4.1 reshape2_1.4.4 futile.options_1.0.1 XML_3.99-0.14
[137] evaluate_0.21 lambda.r_1.2.4 laeken_0.5.2 httpuv_1.6.11
[141] VIM_6.2.2 RANN_2.6.1 tidyr_1.3.0 purrr_1.0.1
[145] polyclip_1.10-4 future_1.32.0 scattermore_1.2 xtable_1.8-4
[149] e1071_1.7-13 RSpectra_0.16-1 later_1.3.1 viridisLite_0.4.2
[153] class_7.3-22 tibble_3.2.1 memoise_2.0.1 AnnotationDbi_1.62.1
[157] cluster_2.1.4 ggplot.multistats_1.0.0 globals_0.16.2
koenvandenberge commented
Hi @sam-israel
Thank you for posting. This indeed does not error, and the fitted values you will get are based on interpolation of the basis functions. It is not something we recommend doing indeed and therefore the function should now return an error as soon as you interpolate to pseudotime values higher than the observed maximum pseudotime.