Example queries

KB-Rank provides a means to conduct text searches regarding protein functions and disease associations associated with protein structures. By default, all text fields that are available for search are selected. On the main page of the tool, one can include or exclude any of the available text fields by checking or unchecking the corresponding selection boxes. To make the selections, users are directed to the advanced text search option at the bottom of the main page of the tool.

Example of an informational search with the word hypertension

One type of search that can be conducted with the KB-Rank tool is an informational search. The user’s goal is to learn more about the structures, functions, and disease associations that relevant to the queried text. As an example, submission of a search with the word hypertension returns into the summary page shown below in Figure 1, panel A. A table shown displays the numbers of protein structural chains returned from annotation sources that organized according to type. These types include small molecule associations, molecular functions, and protein domain assignments. Clicking on one of the resource names yields annotation categories assigned by that resource for the selected annotation type. For example, choosing the DrugBank resource under the annotation type small molecule associations brings the user to the page shown in Figure 1, panel B. Here drugs are ranked according to their relevance to the search term hypertension. The user can learn more about drugs that are used to treat or may potentially be used treat hypertension. Each of these drugs are associated with a protein target with a known three-dimensional structure. The drugs listed include those ranked 2-4 which are enalkiren, aliskiren, and remikiren. Following the corresponding links for these drugs to the DrugBank site provides descriptions the drugs as treatments for hypertension. The drug ranked first, hydroxyfasudil, is not used as current treatment for hypertension in humans. But it has been demonstrated to reduce the rate of bladder dysfunction in hypertensive rats, as described by Inoue et al. (Urology, Feb 24, 2012). The drug may potentially be a future treatment for hypertension in humans.
An alternative selection of annotation categories for the search with the word hypertension is to retrieve information from the EC2PDB resource. That can be done either by clicking the EC2PDB button from the top menu or choosing the EC2PDB link from the summary table shown in panel A. Shown in panel C of Figure 1 are the enzymatic reaction categories that are found to be relevant to hypertension. These include cleavage of Leu-|-Xaa bond in angiotensinogen to generate angiotensin I, EC+; phosphorylation reactions, EC+; and cleavage of the phosphodiesterase bond in cyclic GMP, EC+ Knowledge of these reactions provides the user further information regarding the molecular etiology of hypertension and the reactions to potentially modify for treatment of the disease.

hypertension, summary
hypertension, small molecules
hypertension, enzymes

Example of navigational search with the word melanoma

A second type of search that can be conducted using the KB-Rank tool is a navigational search. Here the goal is to retrieve a structure or structures relevant to the queried text that can be used for further study. A further study may be to use the structure(s) retrieved to foster projects in structure-based drug design.
An example search text is the word melanoma, as shown in Figure 2. By selecting the Structures tab at the top of the report page, the user is directed to a list of structures that are ranked according to their relevance to the query. The highest ranked structure retrieved is the protein dual specificity mitogen-activated protein kinase kinase 1, PDBID+3E8N, chain A. The structure is a known target for the treatment of cancer as described by Wang et al. (Biochem. Biophys. Acta 2007, 1773, 1248). Studies are ongoing for drug development using that structure as described by Montagut et al. (J. Cancer Lett. 2009, 283, 125). The search with the KB-Rank produced a viable protein target for the design of potential new treatments for disease queried, which in this case, is melanoma.

melanoma, structures

If you utilize KB-Rank for your work, please cite the following reference.
Julfayev ES, McLaughlin RJ, Tao YP, McLaughlin WA. KB-Rank: efficient protein structure and functional annotation identification via text query. J Struct Funct Genomics. 2012. DOI: 10.1007/s10969-012-9125-7.