Case Studies
Case study 1 - STRING Database support
Medusa was initially used as a front-end application to support STRING database.
The main advantage of the tools is that it is able to support different types of connections between two nodes.
Such information is provided by STING database and Medusa is a really nice application for such a purpose.
The image her shows a real example of such a network as it is originally produced by STRING database.
Suppose we have determined a link between SYW_PROMA (Tryptophanyl-tRNA synthetase (EC 6.1.1.2) (Tryptophan--tRNA ligase) (TrpRS) in Prochlorococcus marinus; see STRING) and TRPA_THETN (TRPA in Thermoanaerobacter tengcongensis). Let's call this evidence type 'experiments'. Right-press on one of these nodes and create a link to the other by dragging the mouse to the other node. When you release, a dialog will open. Choose 'experiments' from interaction type. Also, change color and shape by right-clicking on the nodes. Data
Case study 2 - Extracting protein complexes from PPI data
Medusa in now a useful tool that goes one step further that visualization.
Hosting a great variety of clustering algorithms, we used Medusa to identify and extract protein complexes from a protein-protein interaction yeast dataset (Gavin, et al., 2006) as presented in (Moschopoulos, et al., 2009).
Medusa was also used as a front-end application to visualize clustering algorithms for jClust clustering platform.
Protein complexes predicted after applying Spectral clustering algorithm and filtering the results in a yeast protein–protein dataset using the jClust application (Pavlopoulos et al. ).The budding yeast Arp2/3 complex that is highlighted was successfully predicted.
Case study 3 - Human-GpDB
Taking advantage of its reach layouts, we used Medusa to visualize workflows of signal transduction from the outer to the inner part of the cell.
The original information was taken from human-gpDB (Satagopam, et al., 2010) database.
Visualization of human Prostanoid TP receptor’s interactions (drugs included).
(a) Arena3D Visualization: Human Prostanoid TP receptor protein of Class A GPCR family targets four Gα G-Proteins that belong to Gq/11 family. The G-Proteins are connected to effectors proteins belonging to eight specific families. For this specific receptor 23 different drugs exist.
(b) Medusa 2D Visualization: Human Prostanoid TP receptor protein targets Gα G-Proteins that belong to Gq/11 family. These G-Proteins interact with 11 different subfamilies of effectors. For this specific receptor 23 different drugs exist.
Case study 4 - Human-GpDB
Taking advantage of its reach layouts, we used Medusa to visualize workflows of signal transduction from the outer to the inner part of the cell.
The original information was taken from human-gpDB (Satagopam, et al., 2010) database.
Visualization of human Prostanoid TP receptor’s interactions (drugs included).
Human postanoid receptors and their interactions. GPCR transmembrane proteins are classified in subfamilies whereas G-proteins and effectors are classified in families according to Human-gpDB database (Satagopam, et al., 2010). A preloaded image of a Cell shows the signal transduction from the outer to the inner part of the cell. DATA
Case study 5 - COAL
Taking advantage of its reach layouts, we used Medusa visualize data coming from COAL application.
In this work, a spectral bipartitioning technique was applied to all proteins from 53 archaeal genomes. Comparisons between different taxonomic levels allowed us to study the effects of phylogenetic distances on cluster structure. Likewise, by associating functional annotations and phenotypic metadata with each protein, we could compare our protein similarity clusters with both protein function and associated phenotype. Our clusters can be analyzed graphically and interactively online.