Hi,
I'm using your DESeq2 package to perform RNAseq analysis of my data, and I would lke to kindly ask you an information about it. In details,here is my R code:
dds <- estimateSizeFactors(dds)
dds <- estimateDispersions(dds)

 When I the estimateDispersions function to my DESeqDataSet object, the following error appears:

"gene-wise dispersion estimates Error in t(hatmatrix %*% t(y)) : error in evaluating the argument 'x' in selecting a method for function 't': Error in hatmatrix %*% t(y) : the arguments are not compatible".

In my first analysis this error didn't  not appear, I don't know how I should solve this problem, and I would like to ask you some suggestions about it.

Here is the output of sessionInfo():

R version 3.1.2 (2014-10-31)
Platform: x86_64-w64-mingw32/x64 (64-bit)

locale:
[1] LC_COLLATE=Italian_Italy.1252  LC_CTYPE=Italian_Italy.1252   
[3] LC_MONETARY=Italian_Italy.1252 LC_NUMERIC=C                  
[5] LC_TIME=Italian_Italy.1252    

attached base packages:
[1] stats4    parallel  stats     graphics  grDevices utils     datasets
[8] methods   base     

other attached packages:
 [1] DESeq2_1.6.3                BiocInstaller_1.16.4       
 [3] limma_3.22.7                geneplotter_1.44.0         
 [5] annotate_1.44.0             XML_3.98-1.1               
 [7] AnnotationDbi_1.28.2        RcppArmadillo_0.4.650.1.1  
 [9] Rcpp_0.11.5                 GenomicRanges_1.18.4       
[11] GenomeInfoDb_1.2.4          IRanges_2.0.1              
[13] S4Vectors_0.4.0             DESeq_1.18.0               
[15] lattice_0.20-30             locfit_1.5-9.1             
[17] Biobase_2.26.0              BiocGenerics_0.12.1        
[19] ConsensusClusterPlus_1.20.0

loaded via a namespace (and not attached):
 [1] acepack_1.3-3.3     base64enc_0.1-2     BatchJobs_1.6      
 [4] BBmisc_1.9          BiocParallel_1.0.3  brew_1.0-6         
 [7] checkmate_1.5.2     cluster_2.0.1       codetools_0.2-11   
[10] colorspace_1.2-6    DBI_0.3.1           digest_0.6.8       
[13] fail_1.2            foreach_1.4.2       foreign_0.8-63     
[16] Formula_1.2-1       genefilter_1.48.1   ggplot2_1.0.1      
[19] grid_3.1.2          gtable_0.1.2        Hmisc_3.15-0       
[22] iterators_1.0.7     latticeExtra_0.6-26 MASS_7.3-40        
[25] munsell_0.4.2       nnet_7.3-9          plyr_1.8.2         
[28] proto_0.3-10        RColorBrewer_1.1-2  reshape2_1.4.1     
[31] rpart_4.1-9         RSQLite_1.0.0       scales_0.2.4       
[34] sendmailR_1.2-1     splines_3.1.2       stringr_0.6.2      
[37] survival_2.38-1     tools_3.1.2         xtable_1.7-4       
[40] XVector_0.6.0

Thank you in advance for your time.

Best regards,

Paola

Can you show the colData(dds) and the design? Did you do anything special in the object construction step?

Thanks Michael. I didn't do anything special.

> colData(dds) 
DataFrame with 12 rows and 2 columns
       condition sizeFactor
        <factor>  <numeric>
LD_FW1     LD_FW  0.7991461
LD_FW2     LD_FW  0.9570691
LD_FW3     LD_FW  1.0044134
LD_FW4     LD_FW  1.1327488
LD_FW5     LD_FW  1.3723747
...          ...        ...
LD_SW2     LD_SW  0.9636662
LD_SW3     LD_SW  0.9319597
LD_SW4     LD_SW  1.0048110
LD_SW5     LD_SW  0.9538346
LD_SW6     LD_SW  1.0527847

Hi Michael,

Rebuilding the dds object and run DESeq without updating bioconductor seems fixed it. I could see that it is not a bug or something. I have no clue why it happened. 

Best,

J.