How to analyze a ratio data set at different replication time point.

Question

[Lila M]

Hi every body, I have a final data set (that I get using repliscan) with ratios a different time point (ES, MS and LS) over the human genome. I would like to analyze the data to see if there are differences in the replication time points

chr start     end   ratio1  ratio2  ratio3  ratioMS ratioLS     window
chr1    0     1000     0    0        0        0         0         0_1000
chr1    1000  2000     0    0        0        0         0         1000_2000
chr1    2000  3000     0    0        0        0         0         2000_3000
chr1    3000  4000  0.5544  0.2558  0.3484    0.39   0.6153       3000_4000
chr1    4000  5000  0.8045  0.6283  0.0203    0.0936 0.4611       4000_5000

I'm trying anova but not sure if is the best aproach... Any idea about how to handle this data? Thanks!

Answer 1

[pbpanigrahi]

Possible things to try out

1. Try find correlation between each time points. Correlation can give you some hint about any possibility of trend/association between time points.
2. Anova you can try, but it will tell you whether time points are different or not? It won't tell you which time points. In that case try pairwise.t.test
3. You can try doing PCA/MDS to see whether time points are clustered in PC space. If not, then this will also give you hint that time points are different.