Cell Maps VNN (Cell Maps Visible Neural Network) Tool
The Cell Maps VNN Tool enables the creation, training, and usage of an interpretable neural network-based models that predict cell response to a drug. A hierarchy in HCX format (for instance, generated by cellmaps_generate_hierarchy tool) is used to define the structure of the visible neural network.
This tool creates an output directories (separately for training, testing/prediction and annotation) where results are stored and registered within Research Object Crates (RO-Crate) using the FAIRSCAPE-cli.
Warning
At the moment, this tool enables a user to perform training and predictions using data formatted for NeST VNN project. This example data is located in the example directory of Cell Maps VNN GitHub Repository.
Overview of Cell Maps VNN Flow
Free software: MIT license
Source code: https://github.com/idekerlab/cellmaps_vnn