DESeq error

[bioBob]

Hi,

I am getting a new (for me) error in DESeq. Perhaps it is just late in the day...

Here is the error:
> ###control vs sick, T
> cds_T_CS<-newCountDataSet(all_T_data_m7[1:33709,], ED_T_m7)
> cds_T_CS = estimateSizeFactors(cds_T_CS)
> cds_T_CS<-estimateDispersions(cds_T_CS,sharingMode="fit-only")
Error in if (nr < 2) stop("nrow(modelMatrix) must be >=2.") :
argument is of length zero

My data looks pretty normal...

> str(all_T_data_m7)
'data.frame': 33714 obs. of 7 variables:
$ L4_R1_ACTTGA_idx8_2T : int 299 30 39 34 18 517 1333 68 98 0 ...
$ L4_R1_AGTTCC_idx14_8T: int 619 53 213 48 46 557 1681 358 282 0 ...

> head(all_T_data_m7)
L4_R1_ACTTGA_idx8_2T L4_R1_AGTTCC_idx14_8T ...
XLOC_000001 299 619 ...
XLOC_000002 30 53 ...
XLOC_000003 39 213 ...

As does my experiment definition data frame...

> ED_T_m7
condition
L4_R1_ACTTGA_idx8_2T sick
L4_R1_AGTTCC_idx14_8T control
L4_R1_CAGATC_idx7_1T sick
L4_R1_CTTGTA_idx12_6T control
L4_R1_GATCAG_idx9_3T sick
L4_R1_GGCTAC_idx11_5T sick
L4_R1_TAGCTT_idx10_4T sick
> str(ED_T_m7)
'data.frame': 7 obs. of 1 variable:
$ condition: Factor w/ 2 levels "control","sick": 2 1 2 1 2 2 2


Any suggestions??

[bioBob]

Well, I figure out what the culprit is, but I am not sure why. It doesn't like the experiment definition data frame. When do this it works:

cds_T_CS<-newCountDataSet(all_T_data_m7[1:33709,], c("sick","control","sick","control","sick","sick","sick"))
cds_T_CS = estimateSizeFactors(cds_T_CS)
cds_T_CS<-estimateDispersions(cds_T_CS,sharingMode="fit-only")