Background: In the graphical modelling of brain data, we are interested in estimating connectivitybetween various regions of interest, and evaluating statistical significance in order to derive a networkmodel. This process involves aggregating results across frequency ranges and several patients, in orderto obtain an overall result that can serve to construct a graph.New method: In this paper, we propose a method based on p-value combiners, which have never beenused in applications to EEG data analysis. This new method is split into two aspects: frequency-wide testsand group-wide tests. The first step can be effectively adjusted to control for false detection rate.Results: This two-step protocol is applied to EEG data collected from distinct groups of mental healthpatients, in order to draw graphical models for each group and highlight structural connectivity differ-ences. Using the method proposed, we show that it is possible to reliably achieve this while effectivelycontrolling for false connections detection.Comparison with existing method(s): Conventionally, the Holm’s Stepdown procedure is used for this typeof problem, as it is robust to type I errors. However, it is known to be conservative and prone to falsenegatives. Furthermore, unlike the proposed methods, it does not directly output a decision rule onwhether to accept or reject a statement.Conclusions: The proposed methodology offers significant improvements over the stepdown procedurein terms of error rate and false negative rate across the network models, as well as in term of applicability.