What is it about?
Positive selection plays an important role in species' evolution. In this paper, we proposed a lightweight CNN (named SweepNet) for identifying selective sweeps and a data preprocessing to further improve accuracy of SweepNet. We evaluated SweepNet's performance by comparing it with other five methods for this task under six different genetic scenarios.
Featured Image
Why is it important?
After investigation, we found that the current summary-statistic-based methods are sensitive to confounding factors and the CNN-based methods have excessive number of parameters which require larger storage space and are prone to overfitting.
Read the Original
This page is a summary of: SweepNet: A Lightweight CNN Architecture for the Classification of Adaptive Genomic Regions, June 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3592979.3593411.
You can read the full text:
Resources
Contributors
The following have contributed to this page
