Filtering by
- All Subjects: Biophysics
- All Subjects: Machine Learning
- Creators: Yan, Hao
The first topic is a Mixed Effects Transfer Learning (METL) model that can flexibly incorporate mixed effects and a general-form covariance matrix to better account for similarity and heterogeneity across subjects. I further develop computationally efficient procedures to handle unknown parameters and large covariance structures. Domain relations, such as domain similarity and domain covariance structure, are automatically quantified in the estimation steps. I demonstrate METL in an application of smartphone-based telemonitoring of PD.
The second topic focuses on an MRI-based transfer learning algorithm for non-invasive surgical guidance of glioblastoma patients. Limited biopsy samples per patient create a challenge to build a patient-specific model for glioblastoma. A transfer learning framework helps to leverage other patient’s knowledge for building a better predictive model. When modeling a target patient, not every patient’s information is helpful. Deciding the subset of other patients from which to transfer information to the modeling of the target patient is an important task to build an accurate predictive model. I define the subset of “transferrable” patients as those who have a positive rCBV-cell density correlation, because a positive correlation is confirmed by imaging theory and the its respective literature.
The last topic is a Privacy-Preserving Positive Transfer Learning (P3TL) model. Although negative transfer has been recognized as an important issue by the transfer learning research community, there is a lack of theoretical studies in evaluating the risk of negative transfer for a transfer learning method and identifying what causes the negative transfer. My work addresses this issue. Driven by the theoretical insights, I extend Bayesian Parameter Transfer (BPT) to a new method, i.e., P3TL. The unique features of P3TL include intelligent selection of patients to transfer in order to avoid negative transfer and maintain patient privacy. These features make P3TL an excellent model for telemonitoring of PD using an At-Home Testing Device.
Molecular engineering is an emerging field that aims to create functional devices for modular purposes, particularly bottom-up design of nano-assemblies using mechanical and chemical methods to perform complex tasks. In this study, we present a novel method for constructing an RNA clamp using circularized RNA and a broccoli aptamer for fluorescence sensing. By designing a circular RNA with the broccoli aptamer and a complementary DNA strand, we created a molecular clamp that can stabilize the aptamer. The broccoli aptamer displays enhanced fluorescence when bound to its ligand, DFHBI-1T. Upon induction with this small molecule, the clamp can exhibit or destroy fluorescence. We demonstrated that we could control the fluorescence of the RNA clamp by introducing different complementary DNA strands, which regulate the level of fluorescence. Additionally, we designed allosteric control by introducing new DNA strands, making the system reversible. We explored the use of mechanical tension to regulate RNA function by attaching a spring-like activity through the RNA clamp to two points on the RNA surface. By adjusting the stiffness of the spring, we could control the tension between the two points and induce reversible conformational changes, effectively turning RNA function on and off. Our approach offers a simple and versatile method for creating RNA clamps with various applications, including RNA detection, regulation, and future nanodevice design. Our findings highlight the crucial role of mechanical forces in regulating RNA function, paving the way for developing new strategies for RNA manipulation, and potentially advancing molecular engineering. Although the current work is ongoing, we provide current progress of both theoretical and experimental calculations based on our findings.