Plexus Connect - Structure Search

    Besides searching in your database by using alphanumeric data fields, you can filter compounds based on their structural features as well. Plexus Connect offers several structure search types for identifying molecules with specific functional groups, compounds similar to each other etc.

    Search type
    Substructure search Finds all structures which contain the query as a substructure.
    Full structure search Finds molecules that are equal to the query structure. No additional fragments or heavy atoms are allowed.
    Similarity search Retrieves compounds structurally similar to the query structure.
    Superstructure search It is the opposite of substructure search: It searches for those target molecules which can be found in the given superstructure query. (In this case the roles of the query and target molecules are simply exchanged, so query properties should be specified on the target!) No query features are allowed on the query side.

    A more detailed summary about each search type can be found in the sections below; while further particulars are provided in the JChem Query Guide.

    In each case, the integrated Marvin JS editor can be used for drawing or importing the query structure. The editor can be launched directly by using the context menu of the "Structure" field (the column header in the case of a table) or in Filters panel c lick on Add new (or images/download/thumbnails/13109276/Plus_icon.png icon) and use fuzzy-search to faster find a structure field . In either case, the Marvin JS editor opens automatically in a dialog window, and you can start building your query structure. Please check the Marvin JS user's guide for more information about drawing chemical structures.


    Substructure search

    When you run a substructure query, hit highlight will be used on the chemical structures of the search results. Hit highlight includes the coloring and alignment of the substructure query inside the search hit molecules. By using the default settings, the substructure query inside the hit molecule will become blue, while the rest of the molecule remains black. Furthermore, the hit molecules will be rotated, so the alignment of the matching part of the molecule will be the same as for the substructure query. (For example, see the coloring of the chlorobenzene moiety in the image below, as well as its alignment according to that of the structure query on the search panel.)


    As a rule, every structural feature of the query molecule ( e.g. , atomic or bond properties, stereochemistry) is taken into consideration during substructure search, i.e. , a query with charge matches only charged targets. At the same time, a query without any charge still matches charged targets. There are, however, some advanced search options available which can make structure search faster and more flexible.

    Tetrahedral stereo Specifies consideration of tetrahedral stereo information during search. When this option is off, central and axial chirality information is ignored. Default value is ON.
    Cis-trans stereo Specifies consideration of cis-trans stereo information during search. When this option is off, double bond, ring cis-trans and cumulene stereo information is ignored. Default value is OFF.
    Include tautomers Specifies turning on or off tautomer search. Default value is OFF.

    The panel containing these advanced options can be opened and closed by clicking on the cogwheel icon. Please note that last set of options used in a search will always be remembered for each user.


    Full structure search

    Full structure search finds those molecules that are equal (in size) to the query structure. By default, special molecular features are taken into account the same way as for substructure search.

    Full structure search uses the same advanced search options as substructure search: Tetrahedral stereo, Cis-trans stereo and Include tautomers .

    Similarity search

    Similarity search finds molecules that are similar to the query structure. Similarity between two molecules can be determined from different molecular descriptors, e.g. , from structural fingerprints.

    Plexus Connect uses the Tanimoto metric for calculating the similarity of two molecules. You can set the acceptable level of similarity by specifying a threshold between 0 and 1, where 0 means that we want to retrieve all molecules from the table, while threshold 1 means that we want to retrieve molecules identical to the query structure. The default threshold value is 0.5.

    In the filtered database view, the search results will be sorted automatically by descending similarity.