HTSview is a graphical user interface (GUI) for the analysis of huge HTS data sets.
|
The tool handles up to 500 000 molecules and allows interactive viewing and browsing of structural data. Due to the fact that the system makes use of caching techniques, it is very memory efficient and can be used on a standard workstation. |
 |
|
The software allows you to view your data in different ways, e.g. sorting by identifiers, classes, activities or clusterings. |
|
It is especially suited for the drawing of large similarity matrices and comes with several meaningful data plots, e.g. enrichment and scatter plots. |
 |
|
The GUI is an extension of the FTrees command line tool and can be used for generation of FTrees scripts for distributed computing via a PVM interface. |
|
HTSview triggers FTrees runs and can then be used for graphically analyzing the FTrees results. |
HTSview is totally interactive in order to give you a quick overview of your data.
|
| application scenarios |
HTSview in combination with FTrees can be used as a virtual screening tool.
|
Fast and fuzzy molecular similarity analyzes based on the FTrees descriptor can be performed. |
|
The software contains a module for the building of multiple ligand models via data fusion which can also be used for virtual screening. |
|
Three different comparison algorithms are incorporated - allowing global and local similarity searches. |
HTSview is ideally suited for doing similarity analysis of small molecules. These can be automatically grouped by clustering.
|
Four different clustering algorithms plus automatic cluster parameter selection methods are implemented. |
 |
|
HTSview comes with a cluster browser for viewing, sorting and selecting cluster neighbours. |
|
An interface to external data mining software for further classification and regression analysis is implemented. |
HTSview performs topological alignments of ligands (2.5D).
|
The algorithms are very fast, aligning several molecules in a matter of seconds. |
 |
|
The alignment methods work for diverse sets of molecules. |
|
Large numbers of multiple comparisons are possible (up to 100). |
|
Each alignment is given a similarity score describing the global and local fit of the structures. |
|
The alignments are used to build multiple ligand models or biophore models (see below). |
|
A new visualization method was developed for fast browsing in multiple alignments. All ligands are embedded in a hexagonal grid - common/similar/matched fragments can easily be highlighted. |
HTSview multiple alignment models can be used to extract preliminary QSAR models.
|
HTSview allows fragment analysis of active and inactive molecules (looking at fragment frequencies). |
 |
 |
|
Biophore models can be built which can be used for predicting the activity of new compounds. A visual interpretation of the SAR hypothesis is supported. |
|
| acknowledgements |
|
HTSview was developed to a large degree in the LeadID project in a cooperation between Aventis Pharma Deutschland GmbH, the Fraunhofer Institute SCAI and BioSolveIT GmbH. The joint R&D project currently being implemented is based on research work that has been started by Prof. Dr. Thomas Lengauer (now Director at the Max-Planck-Institute for Computer Science in Saarbruecken) and Prof. Dr. Matthias Rarey (now Director of the Centre for Bioinformatics at the University of Hamburg) during their time at the institute SCAI. Both scientists are also founders and shareholders of BioSolveIT GmbH.
|
| status & availability |
|
HTSview is currently being tested as prototype software close to beta.
If you would like to evaluate HTSview, or lease the software, please contact us to obtain the package and a license using the license request form.
|
m this link leads back to mother tool FTrees |
