Metadata-Version: 2.4
Name: fasthydromap
Version: 0.1.0
Summary: Predict per-residue Fdewet using SASA + MPNN
Author-email: Samuel Lobo <samuels.lobo@gmail.com>
License-Expression: MIT
Project-URL: Homepage, https://github.com/samlobe/FastHydroMap
Project-URL: Repository, https://github.com/samlobe/FastHydroMap
Project-URL: Issues, https://github.com/samlobe/FastHydroMap/issues
Keywords: protein,hydrophobicity,molecular-dynamics,biophysics,machine-learning
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Requires-Python: <3.14,>=3.11
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy<3.0,>=1.26
Requires-Dist: pandas<3.0,>=2.2
Requires-Dist: mdtraj<2.0,>=1.10
Requires-Dist: pdbfixer
Requires-Dist: openmm<9,>=8.2
Requires-Dist: biopython>=1.85
Requires-Dist: tqdm
Requires-Dist: scipy>=1.11
Requires-Dist: scikit-learn
Requires-Dist: pyyaml
Requires-Dist: setuptools<81
Provides-Extra: torch
Requires-Dist: torch<2.8,>=2.2; extra == "torch"
Provides-Extra: inference
Requires-Dist: torch<2.8,>=2.2; extra == "inference"
Provides-Extra: dev
Requires-Dist: build; extra == "dev"
Requires-Dist: twine; extra == "dev"
Requires-Dist: pytest; extra == "dev"
Requires-Dist: black; extra == "dev"
Requires-Dist: ruff; extra == "dev"
Dynamic: license-file

# FastHydroMap

FastHydroMap predicts per-residue dewetting free energies (`Fdewet`) from protein structures and trajectories.

## Quick Start

Start from a fresh Python environment. FastHydroMap currently supports Python `3.11`, `3.12`, and `3.13`.

```bash
pip install fasthydromap
fasthydromap install-torch
fasthydromap predict your_structure.pdb -o outputs/your_structure_fdewet
```

`fasthydromap install-torch` defaults to the CPU build, which is usually the right choice for current FastHydroMap workloads because SASA preprocessing dominates runtime.

Advanced installation options, Docker usage, GPU Torch variants, and release/developer workflows are documented in [docs/INSTALL.md](/home/sam/Research/FastHydroMap_development/FastHydroMap/docs/INSTALL.md) and [docs/PYPI_RELEASE.md](/home/sam/Research/FastHydroMap_development/FastHydroMap/docs/PYPI_RELEASE.md).

## Inputs

FastHydroMap supports:

- Single protein structures in `PDB` format
- Protein trajectories in `DCD` or `XTC` format together with a matching topology `PDB`

Typical usage:

```bash
# Single structure
fasthydromap predict examples/1A1U.pdb -o outputs/1A1U_fdewet

# Trajectory
fasthydromap predict-trajectory examples/proteinG.pdb examples/proteinG_short.dcd -o outputs/proteinG_fdewet
```

## Outputs

For a single structure, FastHydroMap writes:

- `*.csv`: one row per residue with `Fdewet`; with `--parts`, intrinsic and context columns are included
- `*.pdb`: a copy of the input structure with predicted `Fdewet` written to B-factors

For a trajectory, FastHydroMap writes wide CSV files containing one row per frame and one column per residue.
Use `--parts` to also write intrinsic, context, and per-frame summary CSVs.

## Model Scope

FastHydroMap was trained on structured single-chain proteins and the 20 canonical amino-acid chemistries.
Predictions for PTMs and other non-canonical chemistries should be treated cautiously.

## Test your installation
```bash
fasthydromap predict examples/1A1U.pdb -o outputs/1A1U_fdewet
# Trajectory run (DCD or XTC)
fasthydromap predict-trajectory examples/proteinG.pdb examples/proteinG_short.dcd
```

## Acknowledgements

[Shell Lab](https://theshelllab.org/) and [Shea Group](https://labs.chem.ucsb.edu/shea/joan-emma/)
