Forum:Using Graphics Processing Units (GPUs) in bioinformatics
4
0
Entering edit mode
5.3 years ago
Denis ▴ 310

In our days GPUs have installed on many HPC clusters. I'm using HPC cluster for de novo assembly, SNP calling, reads mapping and other bioinformatic routines. I'm wondering for which bioinformatic tasks/algorithms I may apply GPUs? How can I benefit from using GPUs in my computations?

algorithm GPU • 7.4k views
ADD COMMENT
1
Entering edit mode

Hi,

I have removed the extraneous code commenting you had in your post. Please use these formatting options appropriately. Also, this topic seems more like a forum discussion than a question with well defined "correct" answers, so I will change it to a Forum discussion.

ADD REPLY
3
Entering edit mode
5.3 years ago

Mostly GPUs are allocated to jobs that have molecular modeling involved, as they allows thousands of calculations to be performed in parallel, which these tasks require. People trying to work with protein structures (or structures of other molecules like DNA/RNA) and particle dynamics are likely the ones at your institution who make use of them.

For what you're doing, they aren't going to be any more helpful than just running samples in parallel. Particularly since GPUs are quite expensive (relative to typical processors) and your cluster likely only has so many of them to go around.

ADD COMMENT
3
Entering edit mode
5.3 years ago

GPUs are also used in basecalling of nanopore sequencing data, which use RNN machine learning approaches. Although I haven't looked into that further, I read things about people rewriting tools like aligners to make use of GPUs, but it's still very much a niche area.

ADD COMMENT
1
Entering edit mode

Yeah, this is a good point, machine learning packages will often attempt to make use of them, particularly during model training.

ADD REPLY
0
Entering edit mode

See also https://developer.nvidia.com/Clara-Genomics, and googling "GPU aligner" already returns plenty of hits.

ADD REPLY
2
Entering edit mode
5.0 years ago
RochMesa ▴ 20

Hi...Scalable provider of high performance computing and storage solutions, in cooperation with researchers at the University at Buffalo, announced the introduction of GPU-HMMER, an NVIDIA CUDA implementation and extension of MPI-HMMER. GPU-HMMER and MPI-HMMER are open-source implementations of the HMMER protein sequence analysis suite that profoundly reduce computation times. The MPI-HMMER implementation capitalizes on the computational power of multiple processors on large clusters, whereas GPU-HMMER is designed to leverage NVIDIA GPUs (graphics processing units) to accelerate processing on computing systems. Performance of up to 100x faster than a single core of AMD Shanghai 2.3 GHz has been measured for 3 GPU units.

assembling pcb

ADD COMMENT
1
Entering edit mode
5.3 years ago
ATpoint 85k

HiCCUPS uses GPUs to identify chromosomal loops in HiC data https://github.com/aidenlab/juicer/wiki/HiCCUPS.

ADD COMMENT

Login before adding your answer.

Traffic: 1816 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6