Skip to main content
TSR Research Group

Mirror-Image TSR

Extends TSR with keys that distinguish a triangle from its mirror image, so that stereochemistry is captured rather than lost.

Read the paperComputational Biology and Chemistry, 2023

Two triangles either side of a mirror line, each with vertices labelled aa1, aa2 and aa3 and the same angle theta and MaxDist edge. The triangle on the left is marked "a positive key" and its reflection on the right is marked "a negative key".
A triangle and its reflection receive keys of equal magnitude and opposite sign, which is what allows the two to be told apart.
On this page

Overview

Building on the Triangular Spatial Relationship (TSR) method, Mirror-Image TSR addresses a key limitation: the inability to distinguish between triangles that are mirror images of each other. This distinction is critical, as many biological processes are stereospecific, with molecules such as enzymes and receptors showing preferences for particular geometric arrangements of atoms.

Mirror-Image TSR extends the original approach by incorporating keys that account for the chirality of protein structures. Chirality refers to the geometric property where molecules, despite having the same atomic connectivity, exist as non-superimposable mirror images. This is particularly important in drug development, where enantioselectivity plays a crucial role in determining the function of pharmaceuticals.

By modifying the key generation formula to include mirror-image recognition, Mirror-Image TSR captures the stereochemical properties of protein structures, enabling discrimination between enantiomers. This provides greater precision in identifying dynamic structural changes and molecular recognition processes, making it a useful tool for analysing stereochemistry in proteins and their interactions with ligands and drugs.

Run it with BioTSR

Mirror-image keys are an option of BioTSR's single-structure methods, --mirror. To compute them for chain A of ubiquitin:

bash
python -m pip install "biotsr==1.0.0"
biotsr fetch 1UBQ --pdb-dir structures
biotsr run --method protein --pdb structures/1UBQ.pdb --chain A --mirror --output results

A triangle's key is negated when its angle θ, before it is folded into 0–90°, exceeds 90°, so a triangle and its mirror image receive keys of opposite sign. For ubiquitin chain A the 70,300 triangles give 51,913 distinct keys, 24,909 of them negative, against 45,349 without the option. The files are written under the prefix results/1UBQ_A_protein_mirror, and the category column of the keys file reads mirror.

  • Which methods take it: both protein Cα methods and all 13 atomic methods, from single amino acids to ligands, carbohydrates and nucleotides. The five cross methods refuse it; no earlier cross program had a mirror convention.
  • From Python: biotsr.analyze(..., mirror=True).
  • Comparing results: mirror keys are a different key space from unsigned keys, and BioTSR will not compare the two.
  • Compared with the earlier package: where BioTSR's protein Cα conventions differ from TSR-Package's, such as dividing angles by π rather than 3.14, the TSR method page lists the differences.
Which BioTSR 1.0.0 methods take which option
OptionProtein Cα2 methodsAtomic: amino acids5 methodsAtomic: ligands, carbohydrates, nucleotides8 methodsCross TSR5 methodsEffect on keys
Amino acid grouping--aa-groupingGrouped residue labels: a different key space.
Secondary structure (SSE-TSR)--sseKeys unchanged; each key’s count is split over 18 categories.
Mirror images--mirrorthis pageA key is negated when its angle before folding exceeds 90°: a different key space.
Size filtering--size-filterSame keys; triangles whose longest edge is at or above the cutoff are left out.
Contact cutoff--contact-cutoffChooses which atoms on each side take part; not a key change.
From the capability matrix released with BioTSR 1.0.0. A method refuses an option it does not take, and options combine on one command.

Tutorial 01 runs the option beside the other Cα options, and the user guide, §11 describes every option. For many structures, biotsr batch runs a CSV manifest (tutorial 05).

Source code

  • BioTSR

    Current software

    Used by the examples on this page.

    Python

    Licence Apache License 2.0

  • TSR-Package

    Earlier implementation

    Mirror-image keys were first implemented here, as the mirror_image argument of the TSR() function. It is kept unchanged for provenance and for work that depends on it; its README has its own installation instructions.

    Python

    Licence MIT License