The Biotransport team uses machine-learning approaches and physical models to enhance biotransport for medical and pharmaceutical applications.



Topic 1 – The undergraduate researcher will work on curating data from a collection of journal articles for certain biomolecules. Collecting data would involve (1) Manually collecting data from the literature, (2) writing Python scripts that can identify the location of key molecular descriptors in the text, (3) extracting the values of the molecular descriptors, and (4) evaluating the accuracy of collected data. The collected data will then be used to train a neural network architecture that can predict the relationship between the molecular descriptors and the functionality of molecules. 

Topic 2 - Microrheology has emerged in the past few decades as an effective technique for

understanding the mechanical properties of various materials and systems. This project seeks to develop a form of active microrheology that utilizes a magnetic field to induce the motion of target particles using electromagnets. A successfully designed magnetic microrheology system would be powerful to characterize the viscoelastic properties of biological matter and cells.



  • Machine learning
  • Imaging
  • Microfluidics