Bootstrapping RNA-Seq data

[Sebastiano]

hi all

I have an RNA_Seq dataset (FPKM values) obtained from the ENCODE Caltech datasets for K562 cells.
I have 6 different lists of genes and I want to show that the expression values of my panel of genes is not random...therefore I decided to use R and perform bootstrapping.

bootstrap, for what I know, requires normally distributed data. my RNA-Seq data are not normally distributed and they are strongly skewed on the left.

I used BoxCox transformation

box.cox.powers(RNA_Seq[,2])

to normalize them.

I obtained a negative value (-0.03701075 to be precise) and raised my FPKM values to that. I then used

scale()

to obtain the z-scores and see if the z-scores of my genes are bigger/smaller than +/-1.96.
that's because I am interested to see if my panel of genes is more highly expressed -or it has a lower expression- than what you would expect by chance.

However...and here's the problem...the values with the HIGHEST FPKM became those with the LOWEST z-score and vice versa since I elevated to a negative number...I understand why this happens mathematically, however, I don't really know how to handle the bootstrap now because the data show the exact opposite of what I would reasonably expect!

so...how would you guys handle bootstrap on RNA-Seq data to test my genes of interest?

thanks!!

[Simon Anders]

bootstrap, for what I know, requires normally distributed data.

No, quite the opposite!

If your data is normal or you think it can be transformed to approximate normality with the Box-Cox transformation, then you should use conventional parametric statistics and not the bootstrap.

The whole point of the bootstrap is that it allows to perform inference when you cannot or do not want to make any assumptions about your data's parametric distribution.

Without wanting to sound dismissive: Given that you decided to try a bootstrap for all the wrong reasons, may I suggest you go back to your favorite textbook and refresh your memory before proceeding? ;-)