Xinghu Qin

This paper introduces ALFDA, a machine learning method that improves identification of fine-scale genetic structure and preserves geogenetic distances better than existing methods.

This paper introduces DeepGenomeScan, a deep learning framework that improves detection of genomic regions under natural selection, outperforming traditional methods in identifying loci associated with complex selection patterns.

KLFDAPC improves geographic origin prediction by enhancing population separability and accuracy compared to PCA and DAPC.

The paper assesses information-based summary statistics for spatial genetic structure inference using simulated data and machine learning methods, finding that q=1 diversity metrics and Shannon differentiation measures are most effective.