RP3Net 0.0.1__tar.gz

rp3net-0.0.1/.github/workflows/pypi-publish.yml

@@ -0,0 +1,49 @@
+name: Publish Python distribution to PyPI and TestPyPI
+
+on: 
+  push:
+    tags:
+      - 'v[0-9]+.[0-9]+.[0-9]+'  # Trigger on version tags like v1.0.0
+
+jobs:
+  build:
+    name: Build distribution
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        persist-credentials: false
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: "3.12"
+    - name: Install pypa/build
+      run: python3 -m pip install build --user
+    - name: Build a binary wheel and a source tarball
+      run: python3 -m build
+    - name: Store the distribution packages
+      uses: actions/upload-artifact@v4
+      with:
+        name: python-package-distributions
+        path: dist/
+
+  publish-to-pypi:
+    name: Publish Python distribution to PyPI
+    needs:
+    - build
+    runs-on: ubuntu-latest
+    environment:
+      name: pypi
+      url: https://pypi.org/p/<package-name>  # Replace <package-name> with your PyPI project name
+    permissions:
+      id-token: write  # IMPORTANT: mandatory for trusted publishing
+
+    steps:
+    - name: Download all the dists
+      uses: actions/download-artifact@v4
+      with:
+        name: python-package-distributions
+        path: dist/
+    - name: Publish distribution to PyPI
+      uses: pypa/gh-action-pypi-publish@release/v1

rp3net-0.0.1/.github/workflows/python-app-training.yml

@@ -0,0 +1,29 @@
+# This workflow will install Python dependencies, run tests and lint with a single version of Python
+# For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-python
+
+name: Build and test RP3Net with training dependencies
+
+on:
+  push:
+    branches: [ "main" ]
+  pull_request:
+    branches: [ "main" ]
+
+permissions:
+  contents: read
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: '3.12'
+    - name: Install
+      run: |
+        python -m pip install --upgrade pip
+        pip install .[training]
+    - name: Test with unittest
+      run: python -m unittest discover -s ./tests_training -p '*test.py'

rp3net-0.0.1/.github/workflows/python-app.yml

@@ -0,0 +1,29 @@
+# This workflow will install Python dependencies, run tests and lint with a single version of Python
+# For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-python
+
+name: Build and test RP3Net without training dependencies
+
+on:
+  push:
+    branches: [ "main" ]
+  pull_request:
+    branches: [ "main" ]
+
+permissions:
+  contents: read
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: '3.12'
+    - name: Install
+      run: |
+        python -m pip install --upgrade pip
+        pip install .
+    - name: Test with unittest
+      run: python -m unittest discover -s ./tests -p '*test.py'

rp3net-0.0.1/.github/workflows/testpypi-publish.yml

@@ -0,0 +1,53 @@
+name: Publish Python distribution to PyPI and TestPyPI
+
+on: 
+  push:
+    tags:
+      - 'v[0-9]+.[0-9]+.[0-9]+*'  # Trigger on version tags like v0.0.1rc1
+
+jobs:
+  build:
+    name: Build distribution
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        persist-credentials: false
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: "3.12"
+    - name: Install pypa/build
+      run: python3 -m pip install build --user
+    - name: Build a binary wheel and a source tarball
+      run: python3 -m build
+    - name: Store the distribution packages
+      uses: actions/upload-artifact@v4
+      with:
+        name: python-package-distributions
+        path: dist/
+
+  publish-to-testpypi:
+    name: Publish Python distribution to TestPyPI
+    needs:
+    - build
+    runs-on: ubuntu-latest
+
+    environment:
+      name: testpypi
+      url: https://test.pypi.org/p/<package-name>
+
+    permissions:
+      id-token: write  # IMPORTANT: mandatory for trusted publishing
+
+    steps:
+    - name: Download all the dists
+      uses: actions/download-artifact@v4
+      with:
+        name: python-package-distributions
+        path: dist/
+    - name: Publish distribution to TestPyPI
+      uses: pypa/gh-action-pypi-publish@release/v1
+      with:
+        repository-url: https://test.pypi.org/legacy/

rp3net-0.0.1/.gitignore

@@ -0,0 +1,182 @@
+.scripts
+.vscode
+.venv
+scratch*
+.secrets
+.env
+.nb
+~$*
+.DS_Store
+*log
+.history
+
+# LaTex
+*.aux
+*.fdb_latexmk
+*.fls
+*.synctex.gz
+
+.idea/
+.RData
+
+*.session.sql
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+# .venv
+# env/
+# venv/
+# ENV/
+# env.bak/
+# venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/

rp3net-0.0.1/.vscode/launch.json

@@ -0,0 +1,23 @@
+{
+    // Use IntelliSense to learn about possible attributes.
+    // Hover to view descriptions of existing attributes.
+    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
+    "version": "0.2.0",
+    "configurations": [
+        {
+            "name": "Fit",
+            "type": "debugpy",
+            "request": "launch",
+            "program": "src/RP3Net/rp3_train.py",
+            "console": "integratedTerminal",
+            "justMyCode": false,
+            // "args": ["fit", "-c", "config/trainer_a.yml", 
+            //     "--trainer.default_root_dir", "/tmp/pp",
+            //     "--trainer.accelerator", "gpu",
+            // ],
+            "args": ["fit", "-c", "~/test/rp3/emlc.yml"],
+            "cwd": "${workspaceFolder}"
+        },
+        
+    ]
+}

rp3net-0.0.1/.vscode/settings.json

@@ -0,0 +1,18 @@
+{
+    "python.testing.unittestArgs": [
+        "-v",
+        "-s",
+        ".",
+        "-p",
+        "*_test.py"
+    ],
+    "python.testing.pytestEnabled": false,
+    "python.testing.unittestEnabled": true,
+    "jupyter.notebookFileRoot": "${workspaceFolder}/src",
+    "terminal.integrated.env.linux": {
+        "PYTHONPATH": "${workspaceFolder}/src"
+    },
+    "terminal.integrated.env.osx": {
+        "PYTHONPATH": "${workspaceFolder}/src"
+    },
+}

rp3net-0.0.1/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 tanhevg
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

rp3net-0.0.1/PKG-INFO

@@ -0,0 +1,77 @@
+Metadata-Version: 2.4
+Name: RP3Net
+Version: 0.0.1
+Summary: RP3Net: Modelling of recombinant soluble protein production in E. coli
+Maintainer-email: Evgeny Tankhilevich <evgeny@ebi.ac.uk>
+License: MIT
+Classifier: Development Status :: 3 - Alpha
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
+Requires-Python: <4.0,>=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: torch==2.5.*
+Requires-Dist: pandas
+Requires-Dist: transformers==4.46.*
+Requires-Dist: ml_collections
+Requires-Dist: peft
+Requires-Dist: tqdm
+Provides-Extra: training
+Requires-Dist: lightning[pytorch-extra]==2.4.*; extra == "training"
+Requires-Dist: polars==1.27.*; extra == "training"
+Requires-Dist: wandb; extra == "training"
+Dynamic: license-file
+
+# RP3Net
+RP3Net is an AI model for predicting the results of recombinant small-scale protein production in _E. coli_ from the construct sequence. See [the preprint](https://www.biorxiv.org/content/10.1101/2025.05.13.652824v1) and [supplements](https://ftp.ebi.ac.uk/pub/software/RP3Net/) for more details on how it works.
+
+# Checkpoints
+* https://ftp.ebi.ac.uk/pub/software/RP3Net/v0.1/checkpoints/
+
+# Inference
+## Installation
+```
+pip install RP3Net
+```
+
+## Command line
+Simple usage:
+```
+rp3 -p <path_to_checkpoint_file> -f <in_fasta_file> -o <out_csv_file>
+```
+The `out_csv_file` will contain the dataframe with the ids from the `in_fasta_file` and the predicted probabilities of successfull recombinant small-scale protein production in _E. coli_.
+For more information on the command line arguments, type `rp3 -h`.
+
+## Python interface
+```python
+import RP3Net as rp3
+m = rp3.load_model(rp3.RP3_DEFAULT_CONFIG, '/path/to/checkpoint')
+scores = m.predict(['PRTEINWQENCE', 'PRTEIN', 'SQWENCE'])
+print(scores)
+# tensor([0.4223, 0.4134, 0.4165])
+score_map = m.predict({'seq1': 'PRTEINWQENCE', 'seq2': 'PRTEIN', 'seq3': 'SQWENCE'})
+print(score_map)
+# {'seq1': 0.4223055839538574, 'seq2': 0.41336774826049805, 'seq3': 0.4165498912334442}
+```
+
+The `load_model` function returns the model object that can be used directly for prediction (`predict`), and is otherwise a fully functional implementation of a [Pytorch module](https://pytorch.org/docs/stable/generated/torch.nn.Module.html), so can be used for computing gradients and training as well. The `predict` method accepts either a list of sequences as strings, or a dictionary of sequences keyed by their ids. The return type depends on the input, and is either a one-dimensional tensor or a dictionary of floats. In the former case the order of the scores corresponds to the order of the input sequences, in the latter case the dictionary is keyed by the sequence ids.
+
+## Performance and resource usage
+The command line verstion on a modern CPU (base frequency 2.6 GHz) for a batch of 16 constructs with length under 500aa runs in about 3 minutes, using under 5Gb of RAM.
+
+# Training
+Note that installation for inference does not bring in the libraries that are used for training.
+
+## Installation
+```
+pip install 'RP3Net[training]'
+```
+
+## Command line
+```
+rp3_train fit -c <training_config_file>
+```
+Examples of trainer cofigs can be found under `config` folder. Training is managed by [Pytorch Lightning CLI](https://lightning.ai/docs/pytorch/stable/cli/lightning_cli.html); more information can be found by typing `rp3_train -h`
+
+## Training data
+* https://ftp.ebi.ac.uk/pub/software/RP3Net/v0.1/data/

rp3net-0.0.1/README.md

@@ -0,0 +1,53 @@
+# RP3Net
+RP3Net is an AI model for predicting the results of recombinant small-scale protein production in _E. coli_ from the construct sequence. See [the preprint](https://www.biorxiv.org/content/10.1101/2025.05.13.652824v1) and [supplements](https://ftp.ebi.ac.uk/pub/software/RP3Net/) for more details on how it works.
+
+# Checkpoints
+* https://ftp.ebi.ac.uk/pub/software/RP3Net/v0.1/checkpoints/
+
+# Inference
+## Installation
+```
+pip install RP3Net
+```
+
+## Command line
+Simple usage:
+```
+rp3 -p <path_to_checkpoint_file> -f <in_fasta_file> -o <out_csv_file>
+```
+The `out_csv_file` will contain the dataframe with the ids from the `in_fasta_file` and the predicted probabilities of successfull recombinant small-scale protein production in _E. coli_.
+For more information on the command line arguments, type `rp3 -h`.
+
+## Python interface
+```python
+import RP3Net as rp3
+m = rp3.load_model(rp3.RP3_DEFAULT_CONFIG, '/path/to/checkpoint')
+scores = m.predict(['PRTEINWQENCE', 'PRTEIN', 'SQWENCE'])
+print(scores)
+# tensor([0.4223, 0.4134, 0.4165])
+score_map = m.predict({'seq1': 'PRTEINWQENCE', 'seq2': 'PRTEIN', 'seq3': 'SQWENCE'})
+print(score_map)
+# {'seq1': 0.4223055839538574, 'seq2': 0.41336774826049805, 'seq3': 0.4165498912334442}
+```
+
+The `load_model` function returns the model object that can be used directly for prediction (`predict`), and is otherwise a fully functional implementation of a [Pytorch module](https://pytorch.org/docs/stable/generated/torch.nn.Module.html), so can be used for computing gradients and training as well. The `predict` method accepts either a list of sequences as strings, or a dictionary of sequences keyed by their ids. The return type depends on the input, and is either a one-dimensional tensor or a dictionary of floats. In the former case the order of the scores corresponds to the order of the input sequences, in the latter case the dictionary is keyed by the sequence ids.
+
+## Performance and resource usage
+The command line verstion on a modern CPU (base frequency 2.6 GHz) for a batch of 16 constructs with length under 500aa runs in about 3 minutes, using under 5Gb of RAM.
+
+# Training
+Note that installation for inference does not bring in the libraries that are used for training.
+
+## Installation
+```
+pip install 'RP3Net[training]'
+```
+
+## Command line
+```
+rp3_train fit -c <training_config_file>
+```
+Examples of trainer cofigs can be found under `config` folder. Training is managed by [Pytorch Lightning CLI](https://lightning.ai/docs/pytorch/stable/cli/lightning_cli.html); more information can be found by typing `rp3_train -h`
+
+## Training data
+* https://ftp.ebi.ac.uk/pub/software/RP3Net/v0.1/data/

rp3net-0.0.1/config/trainer_a.yml

@@ -0,0 +1,59 @@
+ckpt_path: last
+seed_everything: 42
+test_after_fit_metric: full_val_auc_roc
+track_metric_checkpoints: best
+data:
+  class_path: RP3Net.training.data.RP3GlobalEmbeddingsLDM
+  init_args:
+    hypers:
+      sources: 
+      - SGC_Stockholm
+      ds_path: ~/rp3/prod/v0.1/data/rp3.csv.gz
+      embeddings_file: ~/data/pp/ai/embeddings/16_01/esm2_t33_650M_UR50D_mean.pt
+      test_val_seed: 42
+      seed: 42
+      training_batch_size: 8
+      val_test_batch_size: 8
+model:
+  class_path: RP3Net.training.lm.RP3LM
+  init_args:
+    hypers:
+      sources: 
+      - SGC_Stockholm
+      model:
+        mode: Training_A
+        fm:
+          type: esm2_650m
+        aggregation: mean
+        classification_head:
+          embedding_dim: 1280
+          bias: false
+          end_bias: true
+          layer_norm: false
+          p_drop: 0.1
+          layers: 
+            d: 1280
+            n: 1
+          nonlinearity: SiLU
+optimizer:
+  class_path: torch.optim.Adam
+  init_args:
+    lr: 1.0e-5
+    weight_decay: 0.0
+trainer:
+  enable_progress_bar: false
+  callbacks:
+  - class_path: lightning.pytorch.callbacks.EarlyStopping
+    init_args:
+      check_on_train_epoch_end: false
+      log_rank_zero_only: true
+      min_delta: 0.0001
+      mode: max
+      monitor: full_val_auc_roc
+      patience: 10
+      stopping_threshold: 0.99
+      strict: true
+      verbose: true
+  - class_path: lightning.pytorch.callbacks.LearningRateMonitor
+    init_args:
+      logging_interval: epoch

rp3net-0.0.1/config/trainer_b.yml

@@ -0,0 +1,66 @@
+ckpt_path: last
+seed_everything: 42
+test_after_fit_metric: full_val_auc_roc
+track_metric_checkpoints: best
+data:
+  class_path: RP3Net.training.data.RP3SequenceEmbeddingsLDM
+  init_args:
+    hypers:
+      sources: 
+      - SGC_Stockholm
+      - SGC_Toronto
+      ds_path: ~/rp3/prod/v0.1/data/rp3.csv.gz
+      embeddings_file: ~/data/pp/ai/embeddings/16_01/esm2_t33_650M_UR50D_unpooled.zip
+      test_val_seed: 42
+      seed: 42
+      training_batch_size: 8
+      val_test_batch_size: 8
+model:
+  class_path: RP3Net.training.lm.RP3LM
+  init_args:
+    hypers:
+      sources: 
+      - SGC_Stockholm
+      - SGC_Toronto
+      model:
+        mode: Training_B
+        fm:
+          type: esm2_650m
+        aggregation: stp
+        stp:
+          seq_dim: 1280
+          d: 256
+          num_heads: 8
+          layer_norm: True
+          p_drop: 0.1
+        classification_head:
+          embedding_dim: 256
+          bias: false
+          end_bias: true
+          layer_norm: false
+          p_drop: 0.1
+          layers: 
+            d: 256
+            n: 1
+          nonlinearity: SiLU
+optimizer:
+  class_path: torch.optim.Adam
+  init_args:
+    lr: 1.0e-5
+    weight_decay: 0.0
+trainer:
+  callbacks:
+  - class_path: lightning.pytorch.callbacks.EarlyStopping
+    init_args:
+      check_on_train_epoch_end: false
+      log_rank_zero_only: true
+      min_delta: 0.0001
+      mode: max
+      monitor: full_val_auc_roc
+      patience: 10
+      stopping_threshold: 0.99
+      strict: true
+      verbose: true
+  - class_path: lightning.pytorch.callbacks.LearningRateMonitor
+    init_args:
+      logging_interval: epoch