Arboreto Gene Regulatory Network Inference FAQs

Question 1

Can Arboreto handle single-cell RNA-seq data?

Accepted Answer

Yes, it is specifically designed to handle the high volume of observations found in single-cell datasets and is a core component of the SCENIC pipeline.

Question 2

Does this skill require a specific environment to run?

Accepted Answer

It requires a Python environment with Dask and Pandas. It is critical to use the 'if __name__ == "__main__":' guard when running scripts to support Dask's multi-processing.

Question 3

How do I scale Arboreto to a computing cluster?

Accepted Answer

You can scale by connecting to a remote Dask scheduler. Simply instantiate a distributed Client with the scheduler's address and pass it to the inference function.

Question 4

What is the main difference between GRNBoost2 and GENIE3?

Accepted Answer

GRNBoost2 uses gradient boosting and is optimized for speed and large datasets, while GENIE3 uses random forests and follows the original regression approach for inference.

Question 5

Is it possible to filter by specific transcription factors?

Accepted Answer

Yes, Arboreto allows you to provide a list of transcription factor names to restrict the regulators considered during the network inference process.

Arboreto Gene Regulatory Network Inference

Key Features

Use Cases

Arboreto Gene Regulatory Network Inference

Key Features

Use Cases