Linh Tran

Year entered the Phd. Program: Fall 2019

Research Advisor: Ryan Gutenkunst

Research Topic: Machine learning approaches for inference of demography and natural selection


Lab web page:

Personal Website:

Research Interests: The goal of my research is to develop better tools for inference from genomic variation data. The rise of modern sequencing technology has given us unprecedented access to genomic data that can reveal the evolutionary histories and future trajectories of natural populations. However, the increasingly large quantity of sequencing data and complexity of evolutionary models present several challenges for existing inference methods. In my dissertation, I aim to develop two novel methods for inference from genomic data based on recent advances in machine learning. The first method will significantly improve the computational efficiency of an existing likelihood-based demographic inference framework. The second method will transform the way genomic data is represented to provide more precise inference of the distribution of fitness effects of new mutations, which is a key parameter for quantifying natural selection.