Accelerating Basecalling with Dataflow Computing

Title

Accelerating Basecalling with Dataflow Computing

 Author

Vermond, Lukas (TU Delft Electrical Engineering, Mathematics and Computer Science; TU Delft Quantum & Computer Engineering)

Contributor

Gaydadjiev, G. (mentor)
Serdijn, W.A. (mentor) ORCID 0000-0003-4973-9677
Popov, M. (mentor) ORCID 0000-0001-7292-5334

 Degree granting institution

Delft University of Technology

 Programme

Computer Engineering

 Date

2020-10-01

 Abstract

In an effort to make DNA sequencing more accessible and affordable, Oxford Nanopore Technologies developed the MinION: A portable cellphone sized DNA sequencing device. Translating the information from this device to a nucleotide sequence is called basecalling, and is done with the aid of artificial neural networks. In this thesis, we accelerate a neural network using the concept of Dataflow programming. The result is a complete basecalling application that relies on an FPGA based platform to run the compute intensive parts. We achieve up to 1.51x speedup over a high-end server with two Intel Xeon Processors, with a rate of 57,040 bases per second. In addition, our implementation uses up to 90.27\% less energy compared to the original implementation.

 Subject

Bio-engineering
Artificial Neural Networks
Basecalling
DNA

 To reference this document use:

http://resolver.tudelft.nl/uuid:09135a70-a066-43df-9530-6f0b0cd5e0fc

 Part of collection

Student theses

 Document type

master thesis

 Rights

© 2020 Lukas Vermond
Files
PDF lukas_vermond_msc_thesis.pdf 1.66 MB
Close viewer