Abstract

The multitiered iterative phasing (MTIP) algorithm is used to determine the biological structures of macromolecules from fluctuation scattering data. It is an iterative algorithm that reconstructs the electron density of the sample by matching the computed fluctuation X-ray scattering data to the external observations, and by simultaneously enforcing constraints in real and Fourier space. This paper presents the first ever MTIP algorithm acceleration efforts on contemporary graphics processing units (GPUs). The Compute Unified Device Architecture (CUDA) programming model is used to accelerate the MTIP algorithm on NVIDIA GPUs. The computational performance of the CUDA-based MTIP algorithm implementation outperforms the CPU-based version by an order of magnitude. Furthermore, the Heterogeneous-Compute Interface for Portability (HIP) runtime APIs are used to demonstrate portability by accelerating the MTIP algorithm across NVIDIA and AMD GPUs.

Highlights

  • The study of structures and functionalities of biological macromolecules plays a vital role in understanding their behavior

  • We ran the Compute Unified Device Architecture (CUDA)-based and the central processing unit (CPU)-based multitiered iterative phasing (MTIP) algorithms for different numbers of radial nodes, and the execution time taken by 96 message passing interface (MPI) ranks for computing the iterative

  • Owing to memory limitations on graphics processing units (GPUs), the CUDA-based MTIP algorithm cannot run more than 95 radial nodes

Read more

Summary

Introduction

The study of structures and functionalities of biological macromolecules plays a vital role in understanding their behavior. Fluctuation X-ray scattering (FXS) (Kam et al, 1981) is an X-ray solution scattering technique used to determine macromolecular structure where multiple identical copies of the sample are exposed to an ultrashort X-ray pulse and the resulting diffraction patterns are collected. By collecting these X-ray snapshots at rates below the rotational diffusion times of the particles, FXS data encode highresolution structural details unlike standard solution scattering techniques such as small-angle X-ray scattering (SAXS), which mainly captures particle size and bulk shape information, and wide-angle X-ray scattering (WAXS), which can give details of hierarchical structure information or ordering within particles.

Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call