When working with large molecular systems, selecting or filtering models based on specific structural criteria can be extremely useful. Whether you need to analyze models with fewer than 100 atoms for performance reasons, or zoom in on structures with a specific number of chains, manually browsing through everything is inefficient and error-prone.
Thankfully, SAMSON’s Node Specification Language (NSL) within the structuralModel attribute space allows users to pinpoint models with exactly the traits you’re looking for. In this guide, we’ll explore how to filter structural models based on their atom-based properties such as the number of atoms, carbon atoms, or residues—saving you time and improving your workflow.
Why Filter by Atom Counts?
Here are some real-world use cases where filtering by atom counts can make your modeling experience significantly more efficient:
- Benchmarking: Choose models with fewer than 1000 atoms to test new simulation setups.
- Conformation analysis: Isolate molecules with a narrow atomic range for precise structural comparison.
- Tutorials or demonstrations: Pick small, representative models suitable for quick rendering or teaching.
Understanding the Syntax
Each structural model attribute follows a basic syntax format: sm.attributeName operator value. Let’s dive into a few commonly used filters.
Number of Atoms
The numberOfAtoms attribute (short name: sm.nat) matches structural models based on their atom count.
Examples:
sm.nat > 100: Matches models with more than 100 atoms.sm.nat 100:200: Matches models with 100 to 200 atoms.
Number of Carbons
The numberOfCarbons attribute (short name: sm.nC) is useful for selecting models with a specific amount of carbon atoms.
Examples:
sm.nC < 10: Models with fewer than 10 carbon atoms.sm.nC 10:20: Models with 10 to 20 carbon atoms.
Number of Residues
Need to identify protein models with detailed sequence information? Use the numberOfResidues attribute (short name: sm.nr).
Examples:
sm.nr > 130: More than 130 residues.sm.nr 100:130: Between 100 and 130 residues.
Number of Chains
To locate models with a specific number of chains, the numberOfChains attribute (sm.nc) is useful.
Examples:
sm.nc < 3: Less than 3 chains.sm.nc 2:4: Between 2 and 4 chains.
Filtering in Practice
All these attributes can be combined within SAMSON’s advanced selection tools. For instance, to filter models that have fewer than 500 atoms but more than 2 chains, you could write:
|
1 |
sm.nat < 500 and sm.nc > 2 |
Using these expressions within your workflows may help you cleanly separate models that are suitable for specific simulations, make visualization easier, or reduce problems related to computational cost.
Whether you are modeling a small molecule library or browsing entire protein complexes, these structural filters will make your dataset more manageable. Learn more by exploring the complete list of attributes in the official documentation.
SAMSON and all SAMSON Extensions are free for non-commercial use. You can get SAMSON here.