Running the ANOVA

We are now ready to run NManova2. This program will loop on aligned nucleosomes (only) and for each pair of aligned nucleosomes it will perform an ANOVA (ANalysis Of VAriance) in order to test if there is any interaction between the two stains and the two experiments (NUC and ChIP). A significant interaction will then mean that for this nucleosome, one strain exhibits hybridization signal in the ChIP experiment while the other one seems to have no signal for that nucleosome. By default, NManova2 will also includes a probe effect into the model.

We will use as input the results of the nucleosome alignment as previously computed by NMalign2.

$NManova2 -a S4/NMalign -o S5/NManova -i NucH3K14_norm.db

By default, the ANOVA will be performed on the whole genome. To parallelize the computation or to just focus on a given chromosome you can ask NManova2 to run on a single chromosome For example, suppose we want to run the analysis only on chromosome chrI:

$NManova2 -a S4/NMalign -o S5/NManova -i NucH3K14_norm.db -s chrI

Now, assuming you ran the ANOVA on the whole genome, you should see on the folder S5/MNabova as many files as the number of chromosomes, each file being prefixed by the corresponding chromosome name.