Quantitative Biology, Statistics, Geometry, Machine Learning

​

Short-term goals

​

I. ML package for biological shape mapping/morphing

 

Through the project, I greatly appreciate the power of ML tools in aiding the analysis of a large amount of messy biological and experimental data. Inspired by the project, I want to develop a machine learning package for mapping. It should require only a little training data, and allow users to automatically map different insect wings to a common wing space. It should be robust for different
types of insect wings, and for biological objects similar to wings, like leaves that also have venation networks. The ML tool should also robustly respect any user-defined constraints of the mapping, such as matching shape boundaries, matching landmark points or matching primary insect veins.

Long term goals

​

I. Apply our methods to other elastoplastic materials

Biological data are often large-scale and noisy. However, if we can analyze the data smartly, we can obtain critical and informative knowledge about the underlying nature of the systems. I want to develop different ML tools for biological data analysis. I hope to cover different aspects of the analysis process, including technical procedures (such as the mapping above), statistical modeling, and statistical inferences. This would be highly beneficial for quantitative biology research, as researchers can then focus on the bigger biological questions, rather than solving the technical challenges.