Mapping the Angiosperm Tree of Life project: Dataset of 100 trees includes 640 taxa with chloroplast, mitochondrial, and ribosomal data. Larger nodes indicate higher node support. Node color shows number of parents, with blue showing small values, orange indicating between 2-8, and red to pink greater than 8. Stephen Smith.


This image shows the real-time monitoring of the performance of a large communications network. The techniques involve actively transmitting packets to receiver nodes and solving an inverse problem (called network tomography) to identify performance degradation. George Michailidis, Vijay Nair, and collaborators.


Inferring Functional Connectivity of Neurons from Multivariate Spike Train Data: Information from the time-stamped firings of multiple neurons is used to infer their functional connectivity using computationally-fast data mining techniques and statistical test procedures. Vijay Nair and collaborators.


Left: Three galaxies --elliptical and S0 galaxies (top); spiral galaxy through a "low-resolution" ground-based telescope (bottom-left) versus the "high resolution" Hubble (bottom-right). Right: 1D elliptically projected light profile with three parameters: size, ellipticity, and power-law of the intensity. Image: Chris Miller.

ARCstack 4-01

The Michigan Institute for Data Science or MINDS is the focal point on the U-M campus for the interdisciplinary study of data science.


Advanced high-throughput workflow processing system for constructing end-to-end computational protocols using heterogeneous datasets, diverse tools, and interactive services. This example shows a complex pipeline workflow for extraction, modeling and analysis of neuroimaging biomarkers analyzed with phenotypic and genetic data.

Figure 1

HIV-1 phylogeny comprising virus samples from 662 patients; 250 terminals are shown. Colors at the terminals of the phylogeny represent the estimated stage of infection of the host. Colors on the interior of the phylogeny represent the estimated stage of infection of the host. Yellow shows lineages that are likely to be outside of the risk group.


Connectogram graphs allow MINDS researchers to investigate associations, causality and develop decision support algorithms for complex multidimensional biomedical data.


Developing webapps for interactive exploration of high-dimensional topology, shape, form and size for diverse array of biomedical objects (e.g., cells, organs, systems).


The graphical PubMed Navigator provides a mechanism to traverse and discover networks of investigators, research groups and scientific concepts.