Through a series of annual workshops focused on professional development in team-based research, skills in data science, interdisciplinary experience, and grant writing, the education program has fostered and supported 11 multi-institutional research teams with members from 15 universities. These teams are led by graduate students and postdocs to form interdisciplinary projects. There is now growing evidence that this these collaborations among our students is resulting in significant results presented at conferences and in journals, while building a community of scholars that will continue interacting in their careers. You can learn more details of these project teams below.
This team is working to elucidate genetic influences on the pathologies and pathways of substance abuse.
The team goal is to design algorithms as well as systems that will defend the large scale distributed machine learning against the known range of adversarial attacks.
This team will develop effective tools to extract clinical value from existing databases by correlating phenotypic, and where possibly genotypic, traits with the administered treatment drugs and their eventual treatment outcomes.
The objective of this student-initiated team is to develop tools to extract and analyze quantitative biological information related to organelle content and localization of target proteins.
The objective of this student-initiated proposal is to develop an automatic quantification algorithm to extract and analyze quantitative biological information related to endocytosis of membrane proteins to facilitate data analysis and reduce user-introduced errors.
We are currently developing an algorithm that will recognize specific cell division defects out of multiple images without the need of human intervention.
This team is working together to apply information theory techniques to accurately automate information gathering from cell biology imaging.
Problems we are looking at include energy in communications and computation with error-correcting codes that attain minimum total communication power, and minimizing energy in computing the Fourier transform.
We will attempt to look at dynamic control goals that guide metabolic function as represented in complex sets of bioinformatics data. Our team will first address how to best analyze gene expression data as a big data problem.
This project was completed in December of 2013. The team brought together machine learning and information theory researchers to estimate models of graph data from network samples.
This project was completed in December of 2013. This team sought to address the problem of improving information theoretic techniques, because almost all current methods rely on measuring phenomena and non-asynmptotic problem formulations, which are poor methods.