Accelerated 2D Classification With ISAC Using GPUs
A widely used approach to analyze single particles in electron microscopy data is 2D classification. This process is very computationally expensive, especially when large data sets are analyzed. In this paper we present GPU ISAC, a newly developed, GPU-accelerated version of the established Iterativ...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-07-01
|
| Series: | Frontiers in Molecular Biosciences |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmolb.2022.919994/full |
| _version_ | 1818119112819212288 |
|---|---|
| author | Fabian Schöenfeld Markus Stabrin Tanvir R. Shaikh Thorsten Wagner Stefan Raunser |
| author_facet | Fabian Schöenfeld Markus Stabrin Tanvir R. Shaikh Thorsten Wagner Stefan Raunser |
| author_sort | Fabian Schöenfeld |
| collection | DOAJ |
| description | A widely used approach to analyze single particles in electron microscopy data is 2D classification. This process is very computationally expensive, especially when large data sets are analyzed. In this paper we present GPU ISAC, a newly developed, GPU-accelerated version of the established Iterative Stable Alignment and Clustering (ISAC) algorithm for 2D images and generating class averages. While the previously existing implementation of ISAC relied on a computer cluster, GPU ISAC enables users to produce high quality 2D class averages from large-scale data sets on a single desktop machine equipped with affordable, consumer-grade GPUs such as Nvidia GeForce GTX 1080 TI cards. With only two such cards GPU ISAC matches the performance of twelve high end cluster nodes and, using high performance GPUs, is able to produce class averages from a million particles in between six to thirteen hours, depending on data set quality and box size. We also show GPU ISAC to scale linearly in all input dimensions, and thereby capable of scaling well with the increasing data load demand of future data sets. Further user experience improvements integrate GPU ISAC seamlessly into the existing SPHIRE GUI, as well as the TranSPHIRE on-the-fly processing pipeline. It is open source and can be downloaded at https://gitlab.gwdg.de/mpi-dortmund/sphire/cuISAC/ |
| first_indexed | 2024-12-11T05:05:02Z |
| format | Article |
| id | doaj.art-8e2ff7e5dbb243a297d3e0d55e1c9083 |
| institution | Directory Open Access Journal |
| issn | 2296-889X |
| language | English |
| last_indexed | 2024-12-11T05:05:02Z |
| publishDate | 2022-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Molecular Biosciences |
| spelling | doaj.art-8e2ff7e5dbb243a297d3e0d55e1c90832022-12-22T01:20:03ZengFrontiers Media S.A.Frontiers in Molecular Biosciences2296-889X2022-07-01910.3389/fmolb.2022.919994919994Accelerated 2D Classification With ISAC Using GPUsFabian SchöenfeldMarkus StabrinTanvir R. ShaikhThorsten WagnerStefan RaunserA widely used approach to analyze single particles in electron microscopy data is 2D classification. This process is very computationally expensive, especially when large data sets are analyzed. In this paper we present GPU ISAC, a newly developed, GPU-accelerated version of the established Iterative Stable Alignment and Clustering (ISAC) algorithm for 2D images and generating class averages. While the previously existing implementation of ISAC relied on a computer cluster, GPU ISAC enables users to produce high quality 2D class averages from large-scale data sets on a single desktop machine equipped with affordable, consumer-grade GPUs such as Nvidia GeForce GTX 1080 TI cards. With only two such cards GPU ISAC matches the performance of twelve high end cluster nodes and, using high performance GPUs, is able to produce class averages from a million particles in between six to thirteen hours, depending on data set quality and box size. We also show GPU ISAC to scale linearly in all input dimensions, and thereby capable of scaling well with the increasing data load demand of future data sets. Further user experience improvements integrate GPU ISAC seamlessly into the existing SPHIRE GUI, as well as the TranSPHIRE on-the-fly processing pipeline. It is open source and can be downloaded at https://gitlab.gwdg.de/mpi-dortmund/sphire/cuISAC/https://www.frontiersin.org/articles/10.3389/fmolb.2022.919994/full2D classificationGPUCUDAcryo-EMSPHIRE2D class averages |
| spellingShingle | Fabian Schöenfeld Markus Stabrin Tanvir R. Shaikh Thorsten Wagner Stefan Raunser Accelerated 2D Classification With ISAC Using GPUs Frontiers in Molecular Biosciences 2D classification GPU CUDA cryo-EM SPHIRE 2D class averages |
| title | Accelerated 2D Classification With ISAC Using GPUs |
| title_full | Accelerated 2D Classification With ISAC Using GPUs |
| title_fullStr | Accelerated 2D Classification With ISAC Using GPUs |
| title_full_unstemmed | Accelerated 2D Classification With ISAC Using GPUs |
| title_short | Accelerated 2D Classification With ISAC Using GPUs |
| title_sort | accelerated 2d classification with isac using gpus |
| topic | 2D classification GPU CUDA cryo-EM SPHIRE 2D class averages |
| url | https://www.frontiersin.org/articles/10.3389/fmolb.2022.919994/full |
| work_keys_str_mv | AT fabianschoenfeld accelerated2dclassificationwithisacusinggpus AT markusstabrin accelerated2dclassificationwithisacusinggpus AT tanvirrshaikh accelerated2dclassificationwithisacusinggpus AT thorstenwagner accelerated2dclassificationwithisacusinggpus AT stefanraunser accelerated2dclassificationwithisacusinggpus |