Fast and Accurate Genomic Prediction Using GPU-Accelerated ML Techniques

EasyChair Preprint 14202

Abstract

The rapid advancements in genomic technologies have revolutionized the field of genomics, enabling researchers to decipher complex genetic information with unprecedented speed and accuracy. Despite these advancements, the computational demands of genomic prediction models have escalated, necessitating more efficient and powerful computational methods. This study explores the integration of GPU-accelerated machine learning (ML) techniques to enhance the performance of genomic prediction models. By leveraging the parallel processing capabilities of GPUs, we achieve significant improvements in the speed and accuracy of predicting genetic traits and disease susceptibility. Our approach involves optimizing ML algorithms for GPU architecture, resulting in reduced computational time and increased predictive accuracy. The proposed GPU-accelerated framework is evaluated on various genomic datasets, demonstrating its efficacy in handling large-scale genomic data and complex prediction tasks. The findings highlight the potential of GPU-accelerated ML techniques to transform genomic research, providing a robust and scalable solution for fast and accurate genomic predictions. This study underscores the importance of computational innovations in genomics, paving the way for more personalized and precise genetic insights in healthcare and research.

Keyphrases: Graphics Processing Units (GPUs), Machine Learning (ML), genomics

@booklet{EasyChair:14202,
  author    = {Abill Robert},
  title     = {Fast and Accurate Genomic Prediction Using GPU-Accelerated ML Techniques},
  howpublished = {EasyChair Preprint 14202},
  year      = {EasyChair, 2024}}