Detecting Overlapping Gene Regions Using the U-Net Attention Mechanism

Author

Samuel Lemma, Minnesota State University, Mankato

Abstract

The current issue of locating, diagnosing, and treating cancer and other diseases linked to specific target genes necessitates the creation of a reliable system for precisely identifying target genes that are initially extracted from a human chromosome. Current methodologies often suffer from overlapping gene regions in the target gene that occurs during the analysis process, which can have a substantial impact on the accuracy of the results. Our recommended approach, which was the appropriate model to apply for this particular problem, is set to enhance the analytical process by utilizing neural networks' U-Net with an attention mechanism. We were able to extract a result with 97.8% Validation accuracy from our proposed model by streamlining the process and generating more precise and timely results.

Advisor

Naseef Mansoor

Committee Member

John Burke

Date of Degree

2023

Language

english

Document Type

APP

Degree

Master of Science (MS)

Program of Study

Data Science

Department

Computer Information Science

College

Science, Engineering and Technology

Recommended Citation

Lemma, S. (2023). Detecting overlapping gene regions using the U-Net attention mechanism [Master’s alternative plan paper, Minnesota State University, Mankato]. Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State University, Mankato. https://cornerstone.lib.mnsu.edu/etds/1381/