0

The more I read the more I get confused.

I have 1 control sample and 1 treated sample. I want to know which one of my genes are expressed higher or lower than others. This is RNA seq data

Since the data is huge, I just made a representative example So the control is 

df1<- structure(list(names = structure(c(1L, 2L, 3L, 4L, 5L, 15L, 8L, 
7L, 10L, 11L, 9L, 12L, 6L, 14L, 13L), .Label = c("A", "AA", "AAA", 
"Aab", "buy", "fdd", "fgd", "gff", "hhk", "kil", "lilk", "lolo", 
"nghg", "vgfh", "yuyn"), class = "factor"), col1A = c(1.64, 0.03, 
0, 4.202, 2.981, 0.055, 0, 0.002, 0.005, 0, 0.002, 0.649, 2.55, 
2.762, 6.402)), class = "data.frame", row.names = c(NA, -15L))

Treated is df2 

df2<- structure(list(names = structure(c(1L, 2L, 3L, 4L, 5L, 15L, 8L, 
7L, 10L, 11L, 9L, 12L, 6L, 14L, 13L), .Label = c("A", "AA", "AAA", 
"Aab", "buy", "fdd", "fgd", "gff", "hhk", "kil", "lilk", "lolo", 
"nghg", "vgfh", "yuyn"), class = "factor"), col1 = c(2.172, 0, 
0, 4.353, 4.581, 0.001, 0.027, 0, 0.002, 0, 0, 0.087, 2.129, 
4.317, 5.849)), class = "data.frame", row.names = c(NA, -15L))

I have read so many posts, I could not really figure which method to use, some say permutation, some say simple t test

Help with a one-sample permutation z-test

How to best visualize one-sample test?

Two independent sample test one sided hypothesis test

Should I look at the samples as the last example?

Can someone comment what direction I should go ?

Learner
  • 789
  • 1
  • 7
  • 16
  • Plotting your data shows me that they are right skewed. Is this the same for your actual data? How many observations do you have per group in your dataset? Maybe edit your question and add histograms (or boxplots) with your actual data. – Stefan Jan 29 '19 at 17:20
  • @Stefan there are over 1000 observation. let me know which test I could do – Learner Jan 29 '19 at 17:36
  • Looks like a regression model is what you need. First you merge the two data sets and add a new categorical variable indicating which set the data belongs to, and then test your hypotheses. – user2974951 Jan 30 '19 at 09:16
  • @user2974951 can you give a solution in R? – Learner Jan 30 '19 at 15:54
  • From the data you provided I see that you have some names (gene names I assume) and also some values. You have data for both a control and a treatment set, so are these rows related then? That is id you perform some sort of test on the same gene twice, a before / after scenario? – user2974951 Jan 31 '19 at 07:52
  • @user2974951 Yes they are related. Row names are the same for both data . the first data is control and second one is treated but independent – Learner Jan 31 '19 at 16:57

0 Answers0