Skip to main content
Fig. 1 | BioData Mining

Fig. 1

From: gammaMAXT: a fast multiple-testing correction algorithm

Fig. 1

classical MaxT versus Van Lishout’s implementation of MaxT. In the classical implementation of MaxT, all T i,j values are computed and stored in memory, \(\forall i=0\dots B\), ∀j=1…m. Then, T i,j is overwritten by T i,j+1 whenever T i,j+1>T i,j , \(\forall i=1\dots B\), ∀j=m−1…1. Finally, p j+1 is overwritten by p j whenever p j >p j +1, ∀j=1…m−1. In Van Lishout’s implementation of MaxT, the [T i,n+1,…,T i,m ] values are computed as before \(\forall i=1\dots B\), but only the maximum values M i are stored in memory (for i>0)

Back to article page