Statistics
GNN#
-
framework: Pytorch Geometric
- Supports Homogeneous and heterogeneous GNNs
- Multiple GNN blocks available
- Explainability pipelines included
- Well-documented
-
LEarning Ressources
- Basics of GNNs : Sanchez-Lengeling et al. A Gentle Introduction to Graph Neural Networks"; Distill 2021
- Bioinformatics Applications : Zhang et al. "Graph Neural Netorks and their current applications in bioinformatics", Front. Genet. 2021
- Network Biology : Zitnil et al. "Current and future directions in network biology" 2023
Shapiro / Wilcoxon / Student#
One common activity in statistics is to test if two independant samples are different between each other.
- First we must check if the distribution follow the Normal distribution with
Shapiro
test. Depending this result we will not apply the same test. - If test is significant => does not follow the Normal distribution, we must then use the
Wilcoxon
test. - If test not significant => the data follow normal distribution, we must use the
Student
test. (N>=20)