Power analysis is essential to optimize the design of RNA-seq experiments and to assess and compare the power to detect differentially expressed genes. PowsimR is a flexible tool to simulate and evaluate differential expression from bulk and especially single-cell RNA-seq data making it suitable for a priori and posterior power analyses.
My input counts frame, counts.true.tmp, has dimensions (12319,30). Running estimateParam() using this frame as the input countData fails with "Error in ParamData$means : $ operator is invalid for atomic vectors". However, if I take a subset of the columns (anything from 3 to 19 columns), the function runs as expected. The problem in this case appears not to be tied to a specific column, i.e. it occurs when I use counts.true.tmp[,c(1:20)] but not if I use counts.true.tmp[,c(1:19)] or counts.true.tmp[,c(2:20)]. Example code:
Because of the above, I initially thought this problem was tied to column number. However, when I used a subset of columns to produce a reproducible example for this issue, I found a different situation: counts.true.tmp[c(1:50),c(1:15,17:30)] works, but counts.true.tmp[c(1:50),c(1:30)] does not, indicating that the problem in this case is specifically with column 16. However, I cannot see anything obviously different about this column, and actually this column gives no problem when we use a smaller subset still, e.g. counts.true.tmp[c(1:50),c(10:16)]. So I'm very confused!
Here's the full counts frame, and session info below.
My input counts frame, counts.true.tmp, has dimensions (12319,30). Running estimateParam() using this frame as the input countData fails with "Error in ParamData$means : $ operator is invalid for atomic vectors". However, if I take a subset of the columns (anything from 3 to 19 columns), the function runs as expected. The problem in this case appears not to be tied to a specific column, i.e. it occurs when I use counts.true.tmp[,c(1:20)] but not if I use counts.true.tmp[,c(1:19)] or counts.true.tmp[,c(2:20)]. Example code:
Because of the above, I initially thought this problem was tied to column number. However, when I used a subset of columns to produce a reproducible example for this issue, I found a different situation: counts.true.tmp[c(1:50),c(1:15,17:30)] works, but counts.true.tmp[c(1:50),c(1:30)] does not, indicating that the problem in this case is specifically with column 16. However, I cannot see anything obviously different about this column, and actually this column gives no problem when we use a smaller subset still, e.g. counts.true.tmp[c(1:50),c(10:16)]. So I'm very confused!
Here's the full counts frame, and session info below.