Entering edit mode
sally_b86
•
0
@sally_b86-13975
Last seen 6.1 years ago
Dear Michael,
In the following coldata I'm obtaining Model matrix not full rank error. I can't remove the batch effect, I want to take it into consideration. As I concluded its replicates and batch that are a linear combination, when I try the same design without one of them it works normally. And I can't remove the replicate effect as well. Any suggestions?
Thank you.
colData:
| replicate | surgeon | batch | time | |
| Ctl.1 | R7 | Bruno | a | t0 |
| Ctl.2 | R8 | Bruno | a | t0 |
| Ctl.3 | R2 | Lionel | d | t0 |
| Ctl.4 | R5 | Bruno | c | t0 |
| Ctl.5 | R6 | Bruno | b | t0 |
| Ctl.6 | R1 | Bruno | f | t0 |
| Ctl.7 | R3 | Lionel | g | t0 |
| Ctl.8 | R4 | Lionel | g | t0 |
| Sham0.1 | R7 | Bruno | a | t1 |
| Sham0.2 | R8 | Bruno | a | t1 |
| Sham0.3 | R6 | Lionel | a | t1 |
| Sham0.4 | R3 | Lionel | c | t1 |
| Sham0.5 | R5 | Bruno | c | t1 |
| Sham0.6 | R4 | Lionel | b | t1 |
| Sham0.7 | R1 | Lionel | e | t1 |
| Sham0.8 | R2 | Bruno | e | t1 |
| Sham45.1 | R8 | Bruno | a | t2 |
| Sham45.2 | R7 | Bruno | a | t2 |
| Sham45.3 | R2 | Lionel | d | t2 |
| Sham45.4 | R1 | Bruno | d | t2 |
| Sham45.5 | R3 | Lionel | c | t2 |
| Sham45.6 | R5 | Bruno | c | t2 |
| Sham45.7 | R4 | Lionel | c | t2 |
| Sham45.8 | R6 | Bruno | b | t2 |
| Sham24.1 | R8 | Bruno | a | t3 |
| Sham24.2 | R7 | Bruno | a | t3 |
| Sham24.3 | R1 | Bruno | d | t3 |
| Sham24.4 | R2 | Lionel | d | t3 |
| Sham24.5 | R5 | Lionel | c | t3 |
| Sham24.6 | R4 | Bruno | c | t3 |
| Sham24.7 | R6 | Lionel | b | t3 |
| Sham24.8 | R3 | Lionel | g |
t3 |
Its a time series experiment to check differential gene expression across time there are many variables like surgeon, biological replicates and library preparation batch that might affect our analysis.
Biological replicates are done in different days.
dds <- DESeqDataSetFromMatrix(countData = cts,
colData = coldata,
design = ~ replicate + batch + surgeon + time)
dds <- DESeq(dds, test = "LRT", reduced = ~ replicate + batch + surgeon )
What does R1 share with R1 at other times? Same Q but batch?
R1 samples represent the mice sacrificed in the same week. I was wondering if I can remove it from the analysis. Unfortunately, when I do PCA plot, it is clear that there's a variation among biological replicates (8 replicates per time point).
I can consider that its a biological variation and not experimental so once taken into consideration it will induce a big variation right?
And what exactly is batch here. What do the batch = a samples have in common?
batches correspond to library preparation, the samples that are prepared together for mRNA enrichment.
sessionInfo()
R version 3.4.1 (2017-06-30)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200)
Matrix products: default
locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C LC_TIME=English_United States.1252
attached base packages:
[1] parallel stats4 stats graphics grDevices utils datasets methods base
other attached packages:
[1] DESeq2_1.16.1 SummarizedExperiment_1.6.5 DelayedArray_0.2.7 matrixStats_0.52.2
[5] Biobase_2.36.2 GenomicRanges_1.28.5 GenomeInfoDb_1.12.2 IRanges_2.10.3
[9] S4Vectors_0.14.5 BiocGenerics_0.22.0
loaded via a namespace (and not attached):
[1] genefilter_1.58.1 locfit_1.5-9.1 splines_3.4.1 lattice_0.20-35
[5] colorspace_1.3-2 htmltools_0.3.6 base64enc_0.1-3 blob_1.1.0
[9] survival_2.41-3 XML_3.98-1.9 rlang_0.1.2 DBI_0.7
[13] foreign_0.8-69 BiocParallel_1.10.1 bit64_0.9-7 RColorBrewer_1.1-2
[17] GenomeInfoDbData_0.99.0 plyr_1.8.4 stringr_1.2.0 zlibbioc_1.22.0
[21] munsell_0.4.3 gtable_0.2.0 htmlwidgets_0.9 memoise_1.1.0
[25] latticeExtra_0.6-28 knitr_1.17 geneplotter_1.54.0 AnnotationDbi_1.38.2
[29] htmlTable_1.9 Rcpp_0.12.13 acepack_1.4.1 xtable_1.8-2
[33] scales_0.5.0 backports_1.1.1 checkmate_1.8.4 Hmisc_4.0-3
[37] annotate_1.54.0 XVector_0.16.0 bit_1.1-12 gridExtra_2.3
[41] ggplot2_2.2.1 digest_0.6.12 stringi_1.1.5 grid_3.4.1
[45] tools_3.4.1 bitops_1.0-6 magrittr_1.5 RSQLite_2.0
[49] lazyeval_0.2.0 RCurl_1.95-4.8 tibble_1.3.4 Formula_1.2-2
[53] cluster_2.0.6 Matrix_1.2-11 data.table_1.10.4 rpart_4.1-11
[57] nnet_7.3-12 compiler_3.4.1