pydssp

library0.9.1pypypi✓ verified 86d ago

PyDSSP is a simplified implementation of the DSSP algorithm for assigning secondary structure to protein structures from 3D coordinates. It supports both PyTorch and NumPy backends. Current version is 0.9.1, with a relatively stable release cadence of a few minor versions per year.

ai-ml

pip install pydssp

github ↗homepage ↗

INSTALL

IMPORT

SIG · PYDSSP

pydssp

ai-mlpythonv0.9.1

harness data pending

Install & Compatibility

Where this runs

No compatibility data collected yet for this library.

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

assign

✓ from pydssp import assign

✗ import pydssp

The main function is 'assign'; importing the module alone doesn't bring assign into scope directly, must use from pydssp import assign.

pydssp.assign

✓ from pydssp import assign; assign(...)

✗ pydssp.assign(...)

If you import pydssp, you must call pydssp.assign, not pydssp.assign.

Assign secondary structure to a protein backbone (N, CA, C atoms) using NumPy backend.

import numpy as np from pydssp import assign # Example coordinates (protein backbone N, CA, C atoms) coords = np.random.randn(10, 3, 3) # 10 residues, 3 atoms, 3 coords secondary = assign(coords, backend='numpy') print(secondary)

Debug

Known issues

gotchaThe input coordinate array shape must be (n_residues, 3, 3) for NumPy, or (batch, n_residues, 3, 3) for PyTorch. A common mistake is to provide (n_residues, 3) or (n_residues, 9).

fix

Ensure coordinates are reshaped to (n_residues, 3, 3) where the last dimension is x,y,z for N, CA, C.

affects: all

gotchaThe backend must be explicitly set to 'numpy' or 'pytorch'. Default is None, which will raise an error.

fix

Pass backend='numpy' or backend='pytorch' to assign().

affects: >=0.9.0

Errors

Common errors & fixes

ValueError: The number of residues must be more than 0

Empty or single-residue input array.

fix

Provide at least 2 residues (n_residues >= 2).

TypeError: assign() missing 1 required positional argument: 'coords'

Calling assign() without arguments or misplacing arguments.

fix

Call assign(coords=your_array, backend='numpy').

Upgrade

Version history

0.9.1latest on PyPI · released Feb 24, 2025

Audit

Dependencies

torchoptionalRequired for PyTorch backend

numpyoptionalRequired for NumPy backend

Agent activity

15 hits · last 30 days

node

Amazon

OpenAI (training)

Resources

githubhttps://github.com/ShintaroMinami/PyDSSP ↗

homepagehttps://github.com/ShintaroMinami/PyDSSP ↗