In this project, various methods for causal inference were applied on a dataset from the paper Causal Protein-Signaling Networks Derived from Multiparameter Single-Cell Data, where they were able to almost reconstruct a known causal signaling pathway using Bayesian networks. The following algorithms were used:
- PC-algorithm
- Tabu search with modified BDe score (bnlearn)
- Greedy interventional equivalence search (GIES) algorithm
- Backshift algorithm
Additionaly, the methods were tested on a simulated dataset, with a similar causal structure as the Sachs data.