Bioinfomatics toolbox hardware optmization?

3 views (last 30 days)
Hi, I am planning to use Matlabs Bioinfomatics toolbox to do RNA-seq analysis from quite a number of biological samples.
The RNA-seq library that I have used has 30 million reads which produces a FASTAQ result of about 3-4GB per sample. Currently the PC I am running Matlab on is a i7 9750H Laptop with 16GB of RAM and a RTX 2070 GPU. For anyone who has worked with the Matlab Bioinfomatics toolbox before, may I know if the low level code has been hardware optmized to run with GPU acceleration? I would need to align 30 million sequences to my reference genome along with QC analysis for each sequence. Would this process be possible to be sped up with GPU? Also, would 16GB of RAM be enough for such a process?
I would appreciate any help and advice I can get on this as I am very new to Matlab itself and I am quite unsure of how the low level code interfaces with my Machine's hardware? i.e. OpenGL, CUDA, etc...

Accepted Answer

Tim DeFreitas
Tim DeFreitas on 10 Feb 2020
The Bioinformatics Toolbox does not have any native alignment functions; for now it relies on Bowtie2, which is available as a support package. To my knowledge Bowtie2 is not optimized to run on GPUs. None of the Bioinformatics Toolbox functions have been tested with or optimized for gpuArrays, but some may benefit from optimizations on the native data types like double.
However, bowtie2 does support multiple threads, and does not have particularly high memory requirement, so 16GB should be enough.
  3 Comments
Tim DeFreitas
Tim DeFreitas on 18 Feb 2020
Yes, bowtie2 is not supported on Windows and will only work on Mac or Linux.
Joshua Soon
Joshua Soon on 18 Feb 2020
If that's the case, then the Bioinfomatics toolbox would not be able to perform alignment if my Matlab was installed in a Window's PC? Or is there a way to change the alignment function to another 3rd party package tool?

Sign in to comment.

More Answers (0)

Categories

Find more on Genomics and Next Generation Sequencing in Help Center and File Exchange

Products


Release

R2019b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!