Research in my group largely revolves around the development and application of bioinfomatics and systemm biological algorithms that we apply on biological/biomedical problems. Furthermore, we focus on the computational analysis of large-scale biological datasets. As for social sciences, research focuses on the analysis and modeling of datasets from social media.
Together with Peter Uetz (Center for the study of biological complexity, Virginia Commonwealth University), we determine and analyse large data sets of interactions between host and pathogen proteins in H. sapiens and E. coli. Furthermore, we determine and analyze interactomes of different bacterial species.
Based on patterns of the emergence of political pressure groups on social media outlets, we develop a theory that will allow us to predict the onset of social uprisings and events of mass mobilization. Furthermore, we apply such approaches to the emergence of extremistic groups in general. Research is carried out with Chaoming Song and Neil Johnson (both Dept. of Physics, University of Miami).
Causal genes &
paths in diseases
We develop algorithms that allows us to find the most causal paths from a disease to a potentially causal gene through a network of molecular interactions in human cells. Specifically, we are applying our approach to patients that suffer from posttraumatic stress disorder (PTSD) and Alzheimers disease (Research is carried out with Amanda Myers, Medical School, University of Miami).
Patterns of group
From social media outlets we collect data about ISIS related groups that we describe using pattern based models Research is carried out with Chaoming Song and Neil Johnson (both Dept. of Physics, University of Miami). Such patterns also play a role in the production of knowledge. Based on a corpus of millions of research pappers, we model the dynamics of group formation to identify factors that make teams of researchers successful. Research is acrried out with Mistunori Ogihara (Dept. of Comp Science, University of Miami and Brian Uzzi, Kellogg School of Business, Northwetern University).
Control of networks
Utilizing molecular interaction networks we develop algorithms that allows us to find a smallsubset of nodes that potentially control a network. In other words, we are interested in finding a subset of proteins that can cover the majority of nodes through their interactions. As a proof of principle, we expect that such nodes have biological and disease significance.