High-performance transformation of protein structure representation from internal to Cartesian coordinates

Area of Science:

• Computational chemistry
• Molecular modeling
• Bioinformatics

Background:

• Converting internal molecular coordinates to Cartesian coordinates is crucial for structural analysis.
• Traditional methods are serial, creating a bottleneck for large polymeric molecules like proteins.

Purpose of the Study:

• To develop a highly parallel algorithm for converting internal coordinates to Cartesian coordinates in polymeric molecules.
• To overcome the inherent linear dependency that necessitates serial processing in traditional methods.

Main Methods:

• A tree-based concatenation of coordinate transforms between molecular segments was employed.
• The algorithm was parallelized efficiently on graphics processing units (GPUs).

Main Results:

• The inherent linear dependency along the polymer chain was successfully removed.
• An order of magnitude speedup was observed using parallel GPU processing compared to serial CPU execution.

Conclusions:

• The presented parallel algorithm offers a significant computational advantage for molecular structure conversions.
• This method is applicable to protein engineering and fitting protein structures to experimental data, enhancing efficiency.