DESeq2: Error: $ operator is invalid for atomic vectors using DESeqDataSetFromMatrix

Question

[Nicolas Rosewick]

Hi,

I want to use DESeq2 to perform some differential expression analysis. I've an expression matrix a2 (raw read count computed using featureCounts) and a design table d2 with two variables (a and b). When I try to create a deseq2 object using DESeqDataSetFromMatrix, an error appears:

> dds <- DESeqDataSetFromMatrix(a2,d2,"~a+b")
Error: $ operator is invalid for atomic vectors

I don't understand why it didn't work. I tried with the pasilla data and it worked perfectly...

Anyone has an idea of the origin of the problem?

Here are a subset of a2 and d2

a2: count Table

a2[1:10,]
                  a_1  a_2  a_3  a_4   a_5  a_6  a_7  a_8 a_9 a_10 a_11 a_12  a_13  a_14  a_15  a_16  a_17 a_18 a_19 a_20  a_21  a_22  a_23  a_24 a_25
ENSG00000006459   752 2883 3088  860  2854 1673 4303 2013   0 4345 1863 4743  2474  5609  5741   893  3559    2 2103 1708  6320  4545  2597  1406 1144
ENSG00000015153   287 1101 1448  607  1470 1748 2130  805   0 1347  717 2186  1014  2402  1714   856  1848    2  484  666  1594  1519   580  1499  548
ENSG00000208892    56  438  418  190   787 1022  712  370   0 1410  447  519   292   509  1432   167   173    1  768  340   520   407   147   229  139
ENSG00000110911   231 1072 1478  681  1094 1073 2035  614   0  653  524  873   799  1669  2815   551  1207    1  630  392  1219  1013   957   531  423
ENSG00000127914  1809 4610 4400 3296 10421 7276 7133 3693   0 9658 4708 9817  5072 14495 14322  4699  4222    5 5790 3964 11958  8508  5991  4912 1621
ENSG00000201367     0   23   25    6    52   42   94   43   0   18   16   92     8    40    17    11    10    0   13   11     8     9     8    20   19
ENSG00000221491    68  361  684  184   465  684  524  196   0  631  208  218   408   584  1908   210   260    2  431  174   469   373   217   368  121
ENSG00000271204    51  488  293  177  1150  696 1183  637   0  627 1186  824   356   423   175   135   148    0  330  173   250   339   156   474  385
ENSG00000067560 17478 9436 4175 9112  7122 4130 5550 2073   0 4584 4954 5485 17002 24242 19505 13329 12473    4 2723 2414 10176 22690 23568 15524 2610
ENSG00000136044   427  709  617  559   826 1253 1098  460   0  689  634 1087   787  1859  1485   469   906    2  651  340  1210  1216   744  1014  442

d2: design table with tow variables

d2
             a   b
a_1    treated old
a_2  untreated old
a_3  untreated old
a_4    treated new
a_5  untreated old
a_6  untreated old
a_7  untreated new
a_8  untreated old
a_9  untreated old
a_10 untreated old
a_11 untreated old
a_12 untreated old
a_13   treated old
a_14 untreated old
a_15 untreated new
a_16   treated new
a_17 untreated old
a_18 untreated old
a_19 untreated old
a_20 untreated old
a_21 untreated old
a_22   treated old
a_23   treated old
a_24 untreated old
a_25 untreated old

sessionInfo()
R version 3.1.3 (2015-03-09)
Platform: i386-w64-mingw32/i386 (32-bit)
Running under: Windows XP (build 2600) Service Pack 3

locale:
[1] LC_COLLATE=French_France.1252  LC_CTYPE=French_France.1252    LC_MONETARY=French_France.1252 LC_NUMERIC=C                   LC_TIME=French_France.1252

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

other attached packages:
 [1] pasilla_0.5.1             DESeq_1.18.0              lattice_0.20-30           locfit_1.5-9.1            Biobase_2.26.0            DESeq2_1.6.3              RcppArmadillo_0.5.100.1.0 Rcpp_0.11.6
 [9] GenomicRanges_1.18.4      GenomeInfoDb_1.2.5        IRanges_2.0.1             S4Vectors_0.4.0           BiocGenerics_0.12.1       BiocInstaller_1.16.4

loaded via a namespace (and not attached):
 [1] acepack_1.3-3.3      annotate_1.44.0      AnnotationDbi_1.28.2 base64enc_0.1-2      BatchJobs_1.6        BBmisc_1.9           BiocParallel_1.0.3   brew_1.0-6           checkmate_1.5.2      cluster_2.0.1        codetools_0.2-10
[12] colorspace_1.2-6     DBI_0.3.1            digest_0.6.8         fail_1.2             foreach_1.4.2        foreign_0.8-63       Formula_1.2-1        genefilter_1.48.1    geneplotter_1.44.0   ggplot2_1.0.1        grid_3.1.3
[23] gridExtra_0.9.1      gtable_0.1.2         Hmisc_3.16-0         iterators_1.0.7      latticeExtra_0.6-26  magrittr_1.5         MASS_7.3-39          munsell_0.4.2        nnet_7.3-9           plyr_1.8.2           proto_0.3-10
[34] RColorBrewer_1.1-2   reshape2_1.4.1       rpart_4.1-9          RSQLite_1.0.0        scales_0.2.4         sendmailR_1.2-1      splines_3.1.3        stringi_0.4-1        stringr_1.0.0        survival_2.38-1      tools_3.1.3
[45] XML_3.98-1.1         xtable_1.7-4         XVector_0.6.0

Answer 1

[Devon Ryan]

Try it without quotes around the design.

[Nicolas Rosewick]

dds <- DESeqDataSetFromMatrix(a2,d2,~a+b)
Warning message:
In DESeqDataSet(se, design = design, ignoreRank) :
  some variables in design formula are characters, converting to factors

[Devon Ryan]

That looks better. The warning is about the dataframe (d2), which apparently is using characters rather than factors.

[Nicolas Rosewick]

thanks it works