ProMod3-A versatile homology modelling toolbox

Abstract

Computational methods for protein structure modelling are routinely used to complement experimental structure determination, thus they help to address a broad spectrum of scientific questions in biomedical research. The most accurate methods today are based on homology modelling, i.e. detecting a homologue to the desired target sequence that can be used as a template for modelling. Here we present a versatile open source homology modelling toolbox as foundation for flexible and computationally efficient modelling workflows. ProMod3 is a fully scriptable software platform that can perform all steps required to generate a protein model by homology. Its modular design aims at fast prototyping of novel algorithms and implementing flexible modelling pipelines. Common modelling tasks, such as loop modelling, sidechain modelling or generating a full protein model by homology, are provided as production ready pipelines, forming the starting point for own developments and enhancements. ProMod3 is the central software component of the widely used SWISS-MODEL web-server.

Fig 1. Geometric description of two stem residues.

Stem residues are represented by N, Cα and C atoms. Parameters are: number of residues in between (l, not shown), Euclidean distance from N-stem C to C-stem N (d) and four angles. The angles α and β describe the direction towards the C-stem N relative to the N-Stem. Accordingly, the angles γ and δ define the direction towards N-stem C relative to C-stem (not shown).

Fig 2. Sidechain modelling benchmark.

Comparison of ProMod3 sidechain modelling performance with SCWRL4 by measuring the fraction of χ₁ angles being within 20° of the reference angles observed in the SCWRL4 test set.

Fig 3. Homology modelling benchmark.

Difference of homology modelling performance between ProMod3 and MODELLER. The same data serves as input for both engines to create models for 190 target sequences. The similarity to the native structure is measured by the lDDT score (red, higher is better) and stereochemistry by the MolProbity score (blue, lower is better).

Fig 4. Custom modelling pipeline.

(A) Probability densities for backbone RMSD (top) /length (bottom) of loop candidates processed in default pipeline (blue, N = 83) and custom pipeline (red, N = 40). (B) Loop candidates from A superposed onto optimal template structure (white, PDB ID: 4B8V). (C) Model built with default (blue) and custom pipeline (red). Spheres mark the stem residues flanking the inserted loops. Both models share the same C-stem (orange) but the N-stems differ. The green wireframe represents the loop from the target structure (PDB ID: 6Q40). (D) Sequence alignment with problematic insertion marked red. The first two sequences represent the target and globally optimal template. The last three represent templates that are globally suboptimal but do not contain the problematic insertion.