Skip to main content
Fig. 3 | BMC Systems Biology

Fig. 3

From: From ERα66 to ERα36: a generic method for validating a prognosis marker of breast tumor progression

Fig. 3

Significance analysis method. Considering the mutual information value for two data vectors, we used a shuffling method on one of these two vectors to estimate the distribution of the mutual information as a random variable. The significance test consists in comparing the obtained value of mutual information for the considered non shuffled data vectors to a function of the standard deviation. Dependence test between two random variables X and Y associated to two gene expressions: By shuffling the data of gene Y (random row permutations), we compared the obtained Mutual Information M (X, Y) results. If the one obtained by using the original computation was significantly high (p-value < a, which is in our case equal to 0.01) w.r.t. the generated ones, we concluded to the dependence of the two variables. Thus we could conclude on the independence hypothesis of the two data vectors. X (green), Y (red): Original data. Y1, Y2, YK: Surrogate data (yellow)

Back to article page