liveBook · Manning

Distance bounds matrix

As explained in Build AI Drug Discovery Pipelines

A distance bounds matrix stores lower and upper allowable distance bounds for pairs of atoms in a molecular structure.

A distance bounds matrix records lower and upper allowable distances between every pair of atoms in a molecule. It gives conformer generation a geometric constraint: a candidate geometry is valid when each pairwise distance stays within its bounds. section 6.1.2

RDKit uses the matrix as part of a distance-geometry workflow:

  • Calculate bounds from the molecule’s connection table and geometric rules. section 6.1.2
  • Smooth the bounds across atom triplets so pairwise distances satisfy the triangle inequality. section 6.1.2
  • Sample a valid random distance matrix, embed it into 3D coordinates, and later optimize the resulting conformer with a force field. section 6.1.2

How Build AI Drug Discovery Pipelines covers it

How deep it goes
Explains bounds, validity, triangle smoothing, random sampling, embedding, and later force-field optimization.
Examples
Uses atom triplets, random distance matrices, and generated molecular conformers as concrete artifacts.
What it assumes
Assumes a molecular connection table and basic geometric reasoning, while defining distance bounds and triangle consistency.
What you can do after
Operate RDKit’s conformer-generation workflow by understanding how distance constraints produce candidate 3D geometries.
How it connects
Connects distance geometry with molecular conformers, triangle inequality, 3D embedding, and force fields.

This explainer was generated by Manning's knowledge graph from the text of Build AI Drug Discovery Pipelines. The section links point to the parts of the book it draws on.