Flow-chart of PubFocus functionality. A) The initial PubFocus search is preformed by using several complementing search menus: basic and advanced menus with various search limits for each, and detailed, allowing experienced users to construct their own search queries. B) Data on relevant search matches is extracted from PubMed in packets of 50 XML-formatted records. C) Impact factor is assigned to each PubMed record from the locally hosted journal ranking database. D) Number of forward citations is established from one of three alternative sources: PubMed Basic, PubMed Central or Google™ Scholar and is assigned to each PubMed record. E) Local temporary database of relevant PubMed records enriched with impact factor and volumetric data on forward citations is created on PubFocus server. F) Publications can be sorted and viewed ranked by either: publication date, first author, last author, impact factor, forward citations or PIF (aka Combined Impact Factor; Fig. 2G). Sorting can be done either in ascending or descending modes. G) Alternatively, statistical analysis on relevant publications can be performed. PubFocus sorts publications by various parameters and provides ranking tables with semi-graphical output. H) Through the use of intuitive limiting tools user can create a focused search query and concentrate statistical analysis on just a subset of initial search results without reinitiating often lengthy data acquisition process. I) Multiple biomedical databases, such as ontologies can be integrated with PubFocus using standard MySQL format. J) MySQL full-text search allows automatic search and extraction of matching terms from titles and abstracts of relevant citations. K) Relevant terms are sorted either based on their occurrence rate or by the build-in ontology categories. L) Matching terms are presented in form of semi-graphical output, allowing selection and search narrowing procedures.