GPU Accelerated Pattern Matching Algorithm for DNA Sequences to Detect Cancer using CUDA

Snehal P. Adey
Department of Computer Engineering and Information Technology, College of Engineering, Pune
College of Engineering, Pune, 2013


   title={GPU Accelerated Pattern Matching Algorithm for DNA Sequences to Detect Cancer using CUDA Dissertation},

   author={Adey, Snehal P},



Download Download (PDF)   View View   Source Source   



Cancer is one of the severe diseases causing one in eight deaths worldwide. It can be cured if detected at the very first stage where the cancer cells stay fixed in their area. In stage two it will start to spread. When it spread to muscles enters in third stage. It may cause organ failure. Last stage is the deadliest and inescapable. Success rate of recovery is highest if cancer is detected in the first stage. Early stage cancer detection can be achieved through genetic tests. The research stated in this paper can be used in Gene testing technology to recognize gene differences by analyzing DNA sequence rather to use complex, expensive equipment provided accurate molecular data is available. The report investigates an efficient and simple mechanism for cancer detection. From the obtained results, an individual is verified whether he/she has chances of getting cancer in future or not though his/her DNA. An ordinary middle class individual may find it prohibitive to use existing diagnosis technology as it is bit expensive. The research is done on GPU using CUDA programming model, accelerating the searching process. This has led significant improvement over serial analysis as it is implemented on GPU. There are around 200 cancer types worldwide that can affect human body.The work has been done to diagnose commonly occurring cancer types. In future if molecular science has brought up new study results about cancer genes for other types of cancer, the work can be extended for all types of cancers.
Rating: 2.8/5. From 2 votes.
Please wait...

* * *

* * *

HGPU group © 2010-2021 hgpu.org

All rights belong to the respective authors

Contact us: