Skip to content

Your First Structure

This page walks you through the complete workflow for generating, validating, and visualising your first protein structure with synth-pdb โ€” from a single command to an interactive 3D view in your browser.

Prerequisites

Make sure synth-pdb is installed:

pip install synth-pdb

For energy minimization you also need OpenMM:

pip install synth-pdb[physics]

Step 1: Generate a Structure

The quickest way to create a structure is by specifying a sequence and a conformation:

synth-pdb --sequence "LKELEKELEKELEKEL" --conformation alpha --output helix.pdb

synth-pdb --sequence "VTVTVTVTVTVTVT" --conformation beta --output sheet.pdb

synth-pdb --length 30 --conformation random --output random.pdb

The output is a full-atom PDB file with backbone + side-chain heavy atoms and hydrogens.

Step 2: Visualise in Your Browser

Add --visualize to open an interactive 3D view immediately after generation:

synth-pdb --sequence "LKELEKELEKELEKEL" --conformation alpha --visualize

This launches a py3Dmol viewer in your default browser, coloured by B-factor (flexibility).

Step 3: Validate the Structure

synth-pdb runs validation automatically and prints a report. To see a detailed quality report:

synth-pdb --sequence "LKELEKELEKELEKEL" --conformation alpha --output helix.pdb --validate

The report checks:

Check What it validates
Bond lengths Engh & Huber Z-scores vs. ideal geometry
Ramachandran ฯ†/ฯˆ angles against Top2018 high-res dataset
Rotamers Side-chain ฯ‡ angles vs. Dunbrack library
Steric clashes van der Waals overlap between non-bonded atoms
Planarity Peptide ฯ‰ angle deviation

Step 4: Energy Minimization (Optional)

For physically refined coordinates, add --minimize:

synth-pdb --sequence "LKELEKELEKELEKEL" --conformation alpha \
  --minimize --output helix_minimized.pdb

This runs OpenMM L-BFGS minimization with implicit OBC2 solvent and the AMBER force field. Typical runtime: 1โ€“5 seconds on CPU.

Step 5: Use the Python API

Everything available on the CLI is also accessible from Python:

from synth_pdb import generate_structure, validate_structure

# Generate
pdb_text = generate_structure(
    sequence="LKELEKELEKELEKEL",
    conformation="alpha",
    minimize=True,
)

# Validate
report = validate_structure(pdb_text)
print(f"Ramachandran favoured: {report['ramachandran_favoured']:.1%}")
print(f"Rotamer outliers:      {report['rotamer_outliers']:.1%}")
print(f"Clashscore:            {report['clashscore']:.1f}")

What's in the PDB File?

ATOM      1  N   LEU A   1       2.345   1.234   0.000  1.00 15.23           N
ATOM      2  CA  LEU A   1       3.812   1.234   0.000  1.00 15.23           C
...
Column Field Notes
7โ€“11 Serial Atom number
13โ€“16 Name Atom name (N, CA, C, O, โ€ฆ)
18โ€“20 Res name 3-letter amino acid code
23โ€“26 Res seq Residue number
31โ€“54 X, Y, Z Cartesian coordinates (ร…)
61โ€“66 B-factor Temperature factor โ€” proxies for local flexibility

The B-factor column in synth-pdb output encodes a physics-derived flexibility profile: high for terminal residues and loop regions, low for the hydrophobic core.

Next Steps