Hi there,
I am currently using DESeq2 and have found some NA values in the p-values. From previous post, it was suggested that it is because of a 0 counts in all sample (therefore missing) and because of the Cook's filtering. In the previous version of DESeq2, we can switch off the filtering using nbinomWaldTest(dds, cooksCutoff=FALSE) and this function was gone it the current version. Upon reading the manual, I was found that I can use the function: replaceOutliersWithTrimmedMean to try and get a p-value instead of NA out of DESeq2. However, I've got the following error message:
Error in matrix(rep(trimBaseMean, ncol(dds)), ncol = dds) :
non-numeric matrix extent
Will it has something to do with the warning message I've got from estimateDispersions(dds)?
gene-wise dispersion estimates
mean-dispersion relationship
final dispersion estimates
Warning messages:
1: In log(ifelse(y == 0, 1, y/mu)) : NaNs produced
2: step size truncated due to divergence
(I have also applied the cqn normalization to my data.)
Thank you.
My session information
R version 3.0.2 (2013-09-25)
Platform: x86_64-unknown-linux-gnu (64-bit)
attached base packages:
[1] parallel splines stats graphics grDevices utils datasets
[8] methods base
other attached packages:
[1] SPIA_2.14.0 KEGGgraph_1.20.0 graph_1.40.1
[4] XML_3.98-1.1 edgeR_3.4.2 limma_3.18.7
[7] DESeq2_1.2.8 RcppArmadillo_0.4.000 Rcpp_0.10.6
[10] GenomicRanges_1.14.4 XVector_0.2.0 IRanges_1.20.6
[13] BiocGenerics_0.8.0 cqn_1.8.0 quantreg_5.05
[16] SparseM_1.03 preprocessCore_1.24.0 nor1mix_1.1-4
[19] mclust_4.2
loaded via a namespace (and not attached):
[1] annotate_1.40.0 AnnotationDbi_1.24.0 Biobase_2.22.0
[4] DBI_0.2-7 genefilter_1.44.0 grid_3.0.2
[7] lattice_0.20-24 locfit_1.5-9.1 RColorBrewer_1.0-5
[10] RSQLite_0.11.4 stats4_3.0.2 survival_2.37-4
[13] tools_3.0.2 xtable_1.7-1
I am currently using DESeq2 and have found some NA values in the p-values. From previous post, it was suggested that it is because of a 0 counts in all sample (therefore missing) and because of the Cook's filtering. In the previous version of DESeq2, we can switch off the filtering using nbinomWaldTest(dds, cooksCutoff=FALSE) and this function was gone it the current version. Upon reading the manual, I was found that I can use the function: replaceOutliersWithTrimmedMean to try and get a p-value instead of NA out of DESeq2. However, I've got the following error message:
Error in matrix(rep(trimBaseMean, ncol(dds)), ncol = dds) :
non-numeric matrix extent
Will it has something to do with the warning message I've got from estimateDispersions(dds)?
gene-wise dispersion estimates
mean-dispersion relationship
final dispersion estimates
Warning messages:
1: In log(ifelse(y == 0, 1, y/mu)) : NaNs produced
2: step size truncated due to divergence
(I have also applied the cqn normalization to my data.)
Thank you.
My session information
R version 3.0.2 (2013-09-25)
Platform: x86_64-unknown-linux-gnu (64-bit)
attached base packages:
[1] parallel splines stats graphics grDevices utils datasets
[8] methods base
other attached packages:
[1] SPIA_2.14.0 KEGGgraph_1.20.0 graph_1.40.1
[4] XML_3.98-1.1 edgeR_3.4.2 limma_3.18.7
[7] DESeq2_1.2.8 RcppArmadillo_0.4.000 Rcpp_0.10.6
[10] GenomicRanges_1.14.4 XVector_0.2.0 IRanges_1.20.6
[13] BiocGenerics_0.8.0 cqn_1.8.0 quantreg_5.05
[16] SparseM_1.03 preprocessCore_1.24.0 nor1mix_1.1-4
[19] mclust_4.2
loaded via a namespace (and not attached):
[1] annotate_1.40.0 AnnotationDbi_1.24.0 Biobase_2.22.0
[4] DBI_0.2-7 genefilter_1.44.0 grid_3.0.2
[7] lattice_0.20-24 locfit_1.5-9.1 RColorBrewer_1.0-5
[10] RSQLite_0.11.4 stats4_3.0.2 survival_2.37-4
[13] tools_3.0.2 xtable_1.7-1
Comment