cellmaps-vnn 0.3.0

Dependencies

• cellmaps_utils

• cellmaps_generate_hierarchy

• ndex2

• optuna

• scikit-learn

• networkx

• pandas

• torch

• torchvision

• torchaudio

Compatibility

• Python 3.8+

Installation

git clone https://github.com/idekerlab/cellmaps_vnn
cd cellmaps_vnn
pip install -r requirements_dev.txt
make dist
pip install dist/cellmaps_vnn*whl

Run make command with no arguments to see other build/deploy options including creation of Docker image

make

Output:

clean                remove all build, test, coverage and Python artifacts
clean-build          remove build artifacts
clean-pyc            remove Python file artifacts
clean-test           remove test and coverage artifacts
lint                 check style with flake8
test                 run tests quickly with the default Python
test-all             run tests on every Python version with tox
coverage             check code coverage quickly with the default Python
docs                 generate Sphinx HTML documentation, including API docs
servedocs            compile the docs watching for changes
testrelease          package and upload a TEST release
release              package and upload a release
dist                 builds source and wheel package
install              install the package to the active Python's site-packages
dockerbuild          build docker image and store in local repository
dockerpush           push image to dockerhub

Before running tests and builds, please install pip install -r requirements_dev.txt

For developers

To deploy development versions of this package

Below are steps to make changes to this code base, deploy, and then run against those changes.

1. Make changes

   Modify code in this repo as desired

2. Build and deploy

# From base directory of this repo cellmaps_vnn
pip uninstall cellmaps_vnn -y ; make clean dist; pip install dist/cellmaps_vnn*wh

Needed files

See the format of the files and examples in examples directory of this repository

• gene2ind.txt: A tab-delimited file where the 1st column is index of genes and the 2nd column is the name of genes.

• cell2ind.txt: A tab-delimited file where the 1st column is index of cells and the 2nd column is the name of cells (genotypes).

• cell2mutation.txt: A comma-delimited file where each row has 718 binary values indicating each gene is mutated (1) or not (0). The column index of each gene should match with those in gene2ind.txt file. The line number should match with the indices of cells in cell2ind.txt file.

• cell2cndeletion.txt: A comma-delimited file where each row has 718 binary values indicating copy number deletion (1) (0 for no copy number deletion).

• cell2amplification.txt: A comma-delimited file where each row has 718 binary values indicating copy number amplification (1) (0 for no copy number amplification).

• training_data.txt: A tab-delimited file containing all data points that you want to use to train the model. The 1st column is identification of cells (genotypes), the 2nd column is a SMILES string of the drug and the 3rd column is an observed drug response in a floating point number, and the 4th column is source where the data was obtained from.

• hierarchy.cx2: Hierarchy in HCX format used to create a visible neural network.

• test_data.txt: A tab-delimited file containing all data points that you want to estimate drug response for. The 1st column is identification of cells (genotypes), the 2nd column is a SMILES string of the drug and the 3rd column is an observed drug response in a floating point number, and the 4th column is source where the data was obtained from.

Usage

For information invoke cellmaps_vnncmd.py -h

The tool can be used in 3 modes: train, predict and annotate.

Example usage

cellmaps_vnncmd.py train ./outdir_training --inputdir examples --config_file examples/config.yaml

cellmaps_vnncmd.py predict ./outdir_prediction --inputdir ./outdir_training --config_file examples/config.yaml

cellmaps_vnncmd.py annotate ./outdir_annotation --model_predictions ./outdir_prediction --ndexuser USERNAME --ndexpassword - --parent_network 0b7b8aee-332f-11ef-9621-005056ae23aa --visibility

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

0.3.0 (2026-07-27)

• Add size attribute to hierarchy nodes indicating the number of genes with data in each node.

• Add functionality to accept input feature files as ro-crate in training and prediction flows.

• Rename model file from model_final.pt to model.pt for consistency.

• Added mode-aware CLI translation with –mode support, RO-Crate validation, and automatic config/model resolution.

• Added config.yaml generation for every training run with used hyperparameters.

0.2.2 (2025-07-25)

• Bug fixes: fixed hyperparameter optimization and generation of config file with optimal parameters

0.2.1 (2025-07-01)

• Bug fixes: fixed _annotate_interactomes_of_systems method’s return value and fix hierarchy annotations

0.2.0 (2025-06-26)

• Add annotation in hierarchy nodes of which gene have data for VNN (train)

• Add fake generator of gene importance scores in interactomes of hierarchy system (the real generator will be implemented in the future)

0.1.0 (2024-12-26)

• First release on PyPI.