DeCypher
          Products    News    Support    Sales    Contact 
  bioinformatics and text mining solutions

Site Search:  

Section Topics:

Introduction to AKS
1. Powerful searching
2. Find specific sentences
3. Highlight relevant text
4. Analyze large document sets
5. Informative ranking
6. Author ranking
7. Focusing searches
8. Manage your project
9. Developer Kit
10. Relate biomedical concepts
11. Knowledge navigation
12. External databases links
Inquire about AKS

News:

Tera-Probe for sensitive, high-throughput oligo searches

Announcing the CodeQuest™ Biocomputing Workstation

GeneMatcher & BlastMachine Replacement Program

DeCypher in the Literature

Newsletter:

Subscribe to the TimeLogic newsletter for product information, special offers and technology whitepapers!

Signup

Send page to a colleague

AKS: Ranking of Relevant Authors

Based on the individual scores of each document, AKS ranks the authors according to their relevance in a specific document set, therefore it is easy to find experts or potential referees in a certain field.

Example
Besides biomedical concetps, people are also interesting. With this example we will see how we can look to a certain field of knowledge, find the relevant authors and check others fields where they have been doing research.

Example 6 Fig 1
Fig 1. List of the most significant authors related to COX1 and COX2 ranked by their relevance among the existing documents that relate to either of the two enzymes

COX-1 and COX-2 are two enzymes very important in pharmacologic research. COX-1 is implicated in homeostatic functions (in stomach and kidney) and COX-2 is implicated in inflammation processes. There are several well-known COX inhibitor drugs like aspirin and active on-going research to develop new COX-2-specific inhibitors that do not interfere with COX-1 activity.

If we are interested in finding the most relevant authors contributing to this area of research AKS can definitely help. AKS analyzes the entire document corpus retrieved by a particular query and extracts a list of authors ranked by relevance on a statistical basis.

Fig. 1 displays a list of the most relevant authors of the documents retrieved after performing a query to search for documents that contain COX-1 or COX-2 (it is a conceptual search, i.e. the documents can contain any of the synonyms for COX-1 and COX-2).

The most relevant author is "DuBois, R N". What does he work on? In which aspects of COX enzymes has he focused his research? What is the relevance of his work? AKS helps to answer these questions as we can retrieve and analyze the set of documents authored by this researcher.

The first step to answer this question is to collect all the documents authored by DuBois, R N. We do a search and we meet the first problem: the name of an author appears in several different ways depending on the journal or even depending on the document itself. We could find this author in several combinations according to the presence or absence of the second name, the usage of initials or the use of intermediate capitals in his last name.

As we are not sure of this author having a unique way of being written, we search for the documents written by any author beginning with "DuBois, R". As we can see in fig. 2, there are several "authors" matching this "name". We reject those with the second initial or name not beginning with N and check the remaining possibilities.

Example 6 Fig 2
Fig 2. Ambiguity resolution for the authors��� names beginning with "DUBOIS, R". We select a set of different ways for naming Raymond Dubois. Learn more about ambiguity resolution at Powerful search capabilities oriented to biomedical needs

Example 6 Fig 4
Fig 3. Result of the combined search of the different possible ways Raymond Dubois is found in the scientific literature.

The set of documents for this author (whatever the "name" used in the document) is displayed in fig. 3.

To take a closer look at DuBois work, we create an analysis of the document set he has written or co-written (fig. 3, for more information on document analysis see Large document set analysis). From this study we can see the chemical substances (fig. 4), genes (fig. 5) and diseases (fig. 6) that are more significantly mentioned in his work. Beginning with the diseases, things are very clear: his work is mainly centered in colorectal cancer. Only marginally (item no. 13) appears "inflammation". In the list of the genes (fig. 5), PTGS2 (also known as COX-2) and PTGS1 (COX-1) are the most significant.

Example 6 Fig 4
Fig 4. Cooccurrences yielded by the analysis (I): list of the most relevant chemicals in the documents authored by Dubois, R N.

Example 6 Fig 5
Fig 5. Cooccurrences yielded by the analysis (II): list of the most relevant genes in the documents authored by Dubois, R N.

Example 6 Fig 6
Fig 6. Cooccurrences yielded by the analysis (I): list of the most relevant diseases in the documents authored by Dubois, R N.

 


The last point is to check the list of chemicals (fig. 4). The three first are related to the chemical substances that are subtrates (arachidonic acid, endoperoxid) and products (eicosanoids) of the cyclooxygenases enzymes. Celecoxib and aspirin are COX inhibitor drugs. In the case of celocoxib the inhibition is specific for COX-2. They have been used to reduce the pain and inflammation in general but what is their relation with colorectal cancer? This question is answered if we check the documents containing these chemicals.

We can focus our search towards the study of the NSAID (Non-steroidal anti-inflammatory drugs) like aspirin or celecoxib in relation with cancer. (for more information on focus searches see Suggestion for focusing searches) We select those terms related with any NSAID in the term list of the analysis result (fig. 7) and launch a new analysis to rank the sentences taking into account the selected terms. This analysis provides a set of meaningful sentences (fig. 8) that shows how DuBois defends the use of NSAIDs in colorectal cancer prevention and cure.

Example 6 Fig 7
Fig 7. Selection of the keywords (terms) related to NSAIDs retrieved by the former analysis to make a second analysis focused in these substances.

Example 6 Fig 8
Fig 8. Sentence list resulting from the second analysis (focused in the NSAID). The sentences are ranked by their relevance for the search and considering the term selection. Note the high informative value of the sentences that are listed. They summarizes the central points of the author's work about the use of NSAIDs to inhibit COX-2 and their role in the prevention and cure of the colorectal cancer.

View the TimeLogic 2006 Products, including AKS (pdf)

Request more information or schedule a demonstration of AKS today!

 

TimeLogic is a brand of Active Motif, Inc. All text © 2006