doc: wip #21
Workflow file for this run
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Design pipeline | |
| on: | |
| push: | |
| jobs: | |
| build: | |
| runs-on: ubuntu-latest | |
| defaults: | |
| run: | |
| shell: bash -el {0} | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - name: Clone GT4SD | |
| run: | | |
| git clone https://github.com/GT4SD/gt4sd-core.git | |
| - uses: conda-incubator/setup-miniconda@v3 | |
| with: | |
| activate-environment: gt4sd | |
| environment-file: gt4sd-core/conda_cpu_linux.yml | |
| python-version: 3.8 | |
| auto-activate-base: false | |
| use-only-tar-bz2: true | |
| - name: Install dependencies | |
| run: | | |
| conda activate gt4sd | |
| cd gt4sd-core | |
| pip install --no-deps . | |
| pip uninstall --yes toxsmi && pip install toxsmi | |
| - name: Load datasets | |
| run: | | |
| conda activate gt4sd | |
| mkdir data | |
| python scripts/load_data.py \ | |
| --uniprot P05067 \ | |
| --affinity_type IC50 \ | |
| --affinity_cutoff 10000 \ | |
| --output_dir data/ \ | |
| --train_size 0.8 \ | |
| --binary_labels | |
| - name: Train toxsmi model | |
| run: | | |
| conda activate gt4sd | |
| python scripts/train_toxsmi.py \ | |
| --train data/train.csv \ | |
| --test data/valid.csv \ | |
| --smi data/mols.smi \ | |
| --language tokenizer \ | |
| --params config/toxsmi_conf.json \ | |
| --model models \ | |
| --name toxsmi_model | |
| - name: Generate molecules with MoLeR | |
| run: | | |
| conda activate gt4sd | |
| head -n 5 data/mols.smi > data/good_docks.smi | |
| python scripts/moler_generate_toxsmi.py \ | |
| --smi_path data/good_docks.smi \ | |
| --param_path config/moler_conf.json \ | |
| --output_path data/moler_filtered \ | |
| --predictor_path models/toxsmi_model/weights/best_F1.pt | |
| - name: Generate more diverse molecules with Regression Transformer | |
| run: | | |
| conda activate gt4sd | |
| python scripts/prepare_rt_data.py \ | |
| --smi_path data/moler_filtered/generated.csv \ | |
| --output_path data/moler_filtered/generated_qed.csv | |
| head -n 10 data/moler_filtered/generated_qed.csv > data/moler_filtered/generated_qed_selected.csv | |
| python scripts/rt_generate.py \ | |
| --smi_path data/moler_filtered/generated_qed_selected.csv \ | |
| --param_path config/rt_conf.json \ | |
| --output_path data/rt | |