1
$\begingroup$

I'm having difficulty normalizing my data.

I was searching for transposable elements in my genome, and after this step, I made counts of reads in some transcripts. I produced something like this:

head(table_tissues_filtered_TE)
                  Lengths ova testes lobe retina suckers brain1 brain2 skin stage
Simple_repeat_80      134  58     77   48     69     115     137  131  195     75
tRNA_1                 59   0     14   12      1      19      12   14   21    104
Simple_repeat_87       26   1     33   12      3      15      24   21   19    180
Simple_repeat_114      22   0      0    0      1       0       0    0    2      7
Simple_repeat_115      30   0      0    0      0       0       0    0    0      1
Simple_repeat_123      22   2      3   317     45      13    652  651   15     21
                              axial                    gland viscera
Simple_repeat_80                 99                       35     557
tRNA_1                            9                        0       3
Simple_repeat_87                  9                        0       4
Simple_repeat_114                 0                      204       0
Simple_repeat_115                 0                       42       0
Simple_repeat_123               333                        5       4

where Lengths are the Length of each elements (simple repeats, etc), and the other columns indicate the reads counted with FeatureCounts. I have another table with the number of reads for each tissue:

head(reads_table)
       ova   testes lobe   retina  suckers   brain1  brain2  skin   stage   axial
      522444 310243 226146  102307  126055   489389  668243  372728 262536  233754
  gland  viscera
  24817   25689

I would like to do an RPkM analysis to normalize the data using R, but I don't know exactly how to do it. Can anyone can help me? thank you!!!

$\endgroup$
0

1 Answer 1

1
$\begingroup$

It's very unlikely that "a RPKM analysis" is the right answer. Assuming you'd like to do differential expression, using tools like DESeq or EdgeR on the count table are likely to be a better thing to do.

For reasons why RPKM is not a good approach, have a read of this answer.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.