Boosting Local Causal Discovery in High-Dimensional Expression Data
This work addresses causal discovery in genomics, offering a more efficient alternative for researchers, but it is incremental as it builds on existing methods like ICP.
The paper tackled the problem of improving Local Causal Discovery (LCD) for predicting causal effects in high-dimensional gene expression data, resulting in an estimator that closely matches the accuracy of the state-of-the-art ICP method while being simpler and more efficient.
We study the performance of Local Causal Discovery (LCD), a simple and efficient constraint-based method for causal discovery, in predicting causal effects in large-scale gene expression data. We construct practical estimators specific to the high-dimensional regime. Inspired by the ICP algorithm, we use an optional preselection method and two different statistical tests. Empirically, the resulting LCD estimator is seen to closely approach the accuracy of ICP, the state-of-the-art method, while it is algorithmically simpler and computationally more efficient.