Artificial intelligence algorithms to detect signals of adaptation from genomic data

Inferring how recent adaptation shaped genomic architecture is crucial to
understanding the molecular mechanisms underpinning diversity among species and
genetic variation within populations. The recent rise of artificial intelligence has
created new opportunities in this area. Novel approaches based on deep learning are
showing great promise, especially for scenarios where existing methods are lacking
power. This project will build on recent advances in the field to further develop deep
learning algorithms to infer adaptive evolution from large-scale genomic data and test
them on data from wild populations and controlled evolve-and-resequence
experiments in Drosophila melanogaster.