In analysis of scientific data, causality is important. Typically in science you aren't really making a model to predict things (although that can be a good way

In analysis of scientific data, causality is important. Typically in science you aren't really making a model to predict things (although that can be a good way to test the model) but rather to understand what is going on, and PGMs are great for that.