Hi,
I am getting an error while plotting dexseq results. I ran dexseq in a very usual way and make good plots but this time I am not sure where and what is wrong with plotting.
Error in plot.window(xlim = c(0, 1), ylim = c(0, max(matr))) :
need finite 'ylim' values In addition: Warning messages:
1: In max(coeff, na.rm = TRUE) :
no non-missing arguments to max; returning -Inf
2: In max(matr) : no non-missing arguments to max; returning -Inf
| sample | run | sex | age | passage | rin | condition | |||
| 1 | Sample_18 | new | M | 64 | 5 | 6.9 | control | ||
| 2 | Sample_206 | new | F | 49 | 2 | 7.4 | control | ||
| 3 | Sample_271 | new | F | 26 | 2 | 8 | control | ||
| 4 | Sample_272 | new | F | 24 | 2 | 6.9 | control | ||
| 5 | Sample_277 | new | M | 23 | 2 | 6.9 | control | ||
| 6 | Sample_53 | new | F | 24 | 2 | 8 | control | ||
| 7 | Sample_16 | new | F | 32 | 2 | 10 | patient | ||
| 8 | Sample_180 | new | M | 22 | 4 | 7.4 | patient | ||
| 9 | Sample_251 | new | M | 22 | 2 | 10 | patient | ||
| 10 | Sample_261 | new | M | 37 | 2 | 9.2 | patient | ||
| 11 | Sample_269 | new | F | 41 | 2 | 7.4 | patient | ||
| 12 | Sample_270 | new | F | 39 | 2 | 9.3 | patient | ||
| 13 | Sample_275 | new | F | 26 | 2 | 9.7 | patient | ||
| 14 | Sample_32 | new | F | 25 | 2 | 8.8 | patient | ||
| 15 | Sample_92 | new | M | 40 | 2 | 9.7 | patient |
fullmodel= ~ sample + exon + exon:age + exon:sex + exon:passage + exon:condition reduced_model = ~ sample + exon + exon:age + exon:sex + exon:passage dxd = DEXSeqDataSet(counts.filt, sampleData= coldata, groupID = gID, featureID = fID, design= fullmodel) dxd = estimateSizeFactors(dxd) dxd = estimateDispersions(dxd, BPPARAM = BP, formula = fullmodel) dxd = testForDEU(dxd, BPPARAM=BPPARAM, fullModel = fullmodel, reducedModel = reduced_model) dxd = estimateExonFoldChanges(dxd, fitExpToVar="condition", BPPARAM = BPPARAM) dxr = DEXSeqResults(dxd)
Following is how my results object look. I got 164 significant DEXs from my data (p < 0.05):
LRT p-value: full vs reduced
DataFrame with 233132 rows and 12 columns
groupID featureID exonBaseMean dispersion stat pvalue padj control patient
<character> <character> <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> <numeric>
ENSG00000000003:001 ENSG00000000003 001 705.8388 0.002036784 0.9374992252 0.3329218 0.7906055 34.01570 34.46083
ENSG00000000003:002 ENSG00000000003 002 329.1962 0.001544407 2.5335265978 0.1114509 0.6094709 24.22142 24.68291
... ... ... ... ... ... ... ... ... ...
ENSG00000273486:001 ENSG00000273486 001 25.77248 NA NA NA NA NA NA
ENSG00000273489:001 ENSG00000273489 001 125.03100 NA NA NA NA NA NA
log2fold_patient_control genomicData countData
<numeric> <GRangesList> <matrix>
ENSG00000000003:001 0.018756457 491:748:798:...
ENSG00000000003:002 0.027229156 229:325:370:...
... ... ... ...
ENSG00000273486:001 NA 24:17:37:...
ENSG00000273489:001 NA 223:275:118:...
I went on to see plotDEXSeq code to debug the error and got a new error:
Error in vst(coeff, object) : less than 'nsub' rows,
it is recommended to use varianceStabilizingTransformation directly
Has anybody every faced this error.
I am also attaching the sessionInfo() below;
R version 3.3.1 (2016-06-21)
Platform: x86_64-apple-darwin13.4.0 (64-bit)
Running under: OS X 10.11.6 (El Capitan)
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] grid parallel stats4 stats graphics grDevices utils datasets methods base
other attached packages:
[1] DEXSeq_1.20.1 RColorBrewer_1.1-2 BiocParallel_1.8.1 splitstackshape_1.4.2
[5] data.table_1.10.0 reshape2_1.4.2 VennDiagram_1.6.17 futile.logger_1.4.3
[9] GenomicFeatures_1.26.2 gridExtra_2.2.1 plyr_1.8.4 tidyr_0.6.1
[13] org.Hs.eg.db_3.4.0 AnnotationDbi_1.36.1 GOSemSim_2.0.4 ggplot2_2.2.1
[17] gplots_3.0.1 DESeq2_1.14.1 SummarizedExperiment_1.4.0 Biobase_2.34.0
[21] GenomicRanges_1.26.2 GenomeInfoDb_1.10.2 IRanges_2.8.1 S4Vectors_0.12.1
[25] BiocGenerics_0.20.0
loaded via a namespace (and not attached):
[1] splines_3.3.1 gtools_3.5.0 Formula_1.2-1 assertthat_0.1
[5] statmod_1.4.27 latticeExtra_0.6-28 Rsamtools_1.26.1 RSQLite_1.1-2
[9] backports_1.0.5 lattice_0.20-34 digest_0.6.11 XVector_0.14.0
[13] checkmate_1.8.2 colorspace_1.3-2 htmltools_0.3.5 Matrix_1.2-8
[17] XML_3.98-1.5 biomaRt_2.30.0 genefilter_1.56.0 zlibbioc_1.20.0
[21] xtable_1.8-2 GO.db_3.4.0 scales_0.4.1 gdata_2.17.0
[25] htmlTable_1.8 tibble_1.2 annotate_1.52.1 nnet_7.3-12
[29] lazyeval_0.2.0 survival_2.40-1 magrittr_1.5 memoise_1.0.0
[33] hwriter_1.3.2 foreign_0.8-67 tools_3.3.1 stringr_1.1.0
[37] munsell_0.4.3 locfit_1.5-9.1 cluster_2.0.5 lambda.r_1.1.9
[41] Biostrings_2.42.1 caTools_1.17.1 RCurl_1.95-4.8 bitops_1.0-6
[45] base64enc_0.1-3 labeling_0.3 gtable_0.2.0 DBI_0.5-1
[49] R6_2.2.0 GenomicAlignments_1.10.0 knitr_1.15.1 dplyr_0.5.0
[53] rtracklayer_1.34.1 Hmisc_4.0-2 futile.options_1.0.0 KernSmooth_2.23-15
[57] stringi_1.1.2 Rcpp_0.12.9 geneplotter_1.52.0 rpart_4.1-10
[61] acepack_1.4.1
Hi
Could you include the code that you used to call plotDEXSeq? Does this happen for all the genes or only for a couple of genes?
Alejandro
Hi Alejandro,
I am experiencing the same error. I don't know if @fizer has noticed this (and this is true for me), that the genomicData column contains no values. Could this be that the reads fail to "map" because of this loss of genomic information?
-Blair
Hi Blair. If you don't specify a gtf file or genomic coordinates when you generate your DEXSeqDataSet object, the genomicData slot will be empty. Did you specify genomic coordinates?
Alejandro