Mofa Tools

Configuration File: remote_tools/mofa_tools.json Tool Type: Remote Tools Count: 1

This page contains all tools defined in the mofa_tools.json configuration file.

Available Tools

run_mofa_factors (Type: RemoteTool)

Run MOFA+ (Multi-Omics Factor Analysis v2; Argelaguet et al., Genome Biology 2020) to integrate m…

run_mofa_factors tool specification

Tool Information:

  • Name: run_mofa_factors

  • Type: RemoteTool

  • Description: Run MOFA+ (Multi-Omics Factor Analysis v2; Argelaguet et al., Genome Biology 2020) to integrate multiple omics ‘views’ measured on the SAME samples into a small set of unsupervised latent factors (analogous to principal components shared across views). Returns the per-sample factor matrix and the fraction of variance each factor explains in each view. Input views are inlined as {view_name: {feature: [value_per_sample]}}; every feature list must have one value per sample, in the same sample order.

Parameters:

  • views (object) (required) Omics views over the SAME samples: {view_name: {feature: [value_per_sample]}}. Every feature list across all views must have the same length = number of samples, in the same sample order.

  • samples (array) (optional) Optional sample names; length must match the per-feature value-list length. Defaults to sample_0..sample_{N-1}.

  • n_factors (integer) (optional) Number of latent factors to infer (default 10).

  • n_iter (integer) (optional) Maximum training iterations (default 1000).

Example Usage:

query = {
    "name": "run_mofa_factors",
    "arguments": {
        "views": "example_value"
    }
}
result = tu.run(query)