mcce-features 0.5.0

Extract MCCE electrostatic features for PClass

MCCE Electrostatic Features

This project is to extract electrostatic features from a protein structure. Proposed stages are:

• Stage 1. Extract proposed electrostatic features an MCCE working directory.
• Stage 2. Extract electrostatic features from a pdb file directly from MCCE-ML module.
• Stage 3. Integrate electrostatic feature to PClass classifier

The proposed electrostatic features are:

Core Composition (5)

• net_charge
• isoelectric_point
• acid_fraction_all_residues
• base_fraction_all_residues
• acid_to_base_ratio

pKa perturbation (6)

• acid_big_pka_shift_fraction_all_residues
• base_big_pka_shift_fraction_all_residues
• acid_big_pka_shift_fraction_acids_only
• base_big_pka_shift_fraction_bases_only
• mean_abs_pka_shift
• max_abs_pka_shift

Surface charge (4)

• surface_net_charge
• surface_acid_to_base_ratio
• surface_positive_charge_density
• surface_negative_charge_density

Patch localization (6)

• largest_positive_patch_area
• largest_negative_patch_area
• largest_positive_patch_charge
• largest_negative_patch_charge
• largest_positive_patch_density
• largest_negative_patch_density

Charge Assymetry (4)

• all_charge_dipole_magnitude
• surface_charge_dipole_magnitude
• all_charge_dipole_normalized
• surface_charge_dipole_normalized

Quick Start

Installation

1. Clone the repository

git clone https://github.com/pclass-lab/elefeatures.git

2. Install in Editable (Development) Mode Editable installs allow you to modify the code and immediately test changes.

pip install -e .

After this, the CLI entry point will be available:

mcce-features

To extract features from an MCCE folder

mcce-features extract <folder_name>