How to use Biomine Explorer
The Biomine Explorer offers a search interface and an advanced graph visualization and exploration interface. A typical procedure for using Biomine Explorer consists of the following steps:
- Select the database (network);
- Input query terms one by one using the input field or import several terms at once using the import button;
- Add query terms to the source or target query (or both) using the provided buttons "add to source" and add to target";
- Adjust other search parameters if needed;
- Click the "Search" button and wait for the preview;
- Click the "advanced view" button to open the results in the Biomine Explorer graph visualization interface.
The search interface
The search interface offers the following options:
- Term input field which supports powerful autosuggest. If at least two characters are provided, the input field will present a list of results which can be browsed and selected.
- Import button which can import several terms at once (space or comma separated). The import function will try its best to resolve the provided names. Short names and standard identifiers (e.g. BRCA1, EntrezGene:37916, GO:0000185, etc.) can be used and they are automatically resolved to full Biomine Explorer identifiers. The imported terms can be added to the source or the target query set. The import is case sensitive!
- Database selector offers a list of stable databases (the databases are actually networks but are stored in a database). Typically, there is one available for each organism set (biomine, human and plants).
- Network size selector offers the possiblity to loosely control the size of the resulting network. The values range from "tiny" to "XXX large" which translates to 10 and 200, respectively. Please note that the size constraint is a weak constraint and there is no guarantee that the resulting network will have the desired number of nodes.
- Grouping of edges allows to group all nodes that have the same edges into one node thus reducing the network but preserving the structure. For example, 100 Pubmed article nodes with the same edges will be grouped into a node "Article_(PubMed)_x100".
Biomine Explorer supports two search modes: neigbourhood and connection search. They are used automatically according to which input term sets are provided.
In the neighborhood mode the results is a subnetwork containing all nodes that can be reached from any input node by a path (that fulfills the threshold). If you want to use this mode, add your query terms only to the "source" query term list.
The connection mode works with one or two query sets.
- If there is one (target) query set, the result is a subnetwork containing all nodes that belong to any path (that fulfills the threshold) between any pair of input nodes. If you want to use this mode, add your query terms to the "target" query term list.
- If there are two query sets, the result is a subnetwork containing all nodes that belong to any path (that fulfills the threshold) from any node in the first set to any node in the second set. If you want to use this mode, partition your query terms to the "source" and the "target" query term list.
The network visualization interface
The Biomine Explorer network visualization interface features visualization, exploration, editing and management options. All types of nodes and edges can be deleted using the provided select buttons. Isolated nodes can also be removed. Node or set of nodes can be expanded which means that the search will be called using the selected nodes and the results will be merged into the existing network.
The urls in the visualization interface are static which means that they can be included in scientific publications or published on the web. Moreover, registered users can also save their modified graphs including positions of nodes. This can be very useful for manually improving the default force-based graph layout for specific needs. It is also possible to export and download the network in several popular formats. Private networks can be renamed, cloned, deleted, and made public or private.