MedScan: Information Harvesting
High-throughput data generation methodologies like microarray gene expression require new approaches for gathering information for data analysis. For the best results, computational approaches used for high-throughput data analysis require that biological information from the literature be a coherent and integrated part of the analysis software itself.
Pathway Studio meets this challenge through its MedScan Technology and underlying ResNet database. All editions of Pathway Studio contain MedScan Technology to harvest information from the literature and to save this information in the Pathway Studio ResNet database ready for data analysis.
MedScan is more than a web search engine. Indeed, the output of a Google search can be channeled into MedScan for example. Web searches, like Google, are excellent at finding items as a result of a query. A quick look at the output list usually locates the item for which you are looking. This approach however, is not well suited for information and knowledge gathering. Also, once information is gathered, where do you put it for later computational use? MedScan meets this challenge for the area of biomedical literature and biomedical online information.
PubMed meets the needs for a central repository of biomedical literature. Researchers can go to PubMed and search for any topic and articles of interest, much like a web type of search. However, just like a web type of search, PubMed also provides a list of all the hits with a link to the articles. If a single article, or even just a few, are sought, this search approach is useful. Alternatively, MedScan will list all the articles of interest but additionally scans the text for "relationships", highlights these relationships in the articles and then lists these relationships and the biological molecules and processes involved in the relationships in separate tables. The tables of relationships can be viewed graphically in Pathway Studio and can be saved into the ResNet database for use in experimental data analysis.
Pathway Studio powered by MedScan therefore can be assigned to search all literature sources that you have available to you online. For example, to create and maintain the ResNet Databases, Ariadne uses Pathway Studio Enterprise with MedScan Enterprise. MedScan reads the entire PubMed and 47 full-text journals. You may have access to additional journals and also to additional internal information and sources.
While it takes Ariadne several days to build the ResNet database we provide ResNet with Pathway Studio to save you this trouble. However, as mentioned there may be additional information that you may need in your database and you may wish to make sure this information is as up-to-date as possible. MedScan provides you this functionality.
How good is MedScan?
The best testament to the ability of MedScan to provide coverage of all the facts in the literature with few false positives, is in the number of researchers that use the software and have published in peer review journals. We can't tell you all of our customers but we can use MedScan to search in the literature for "Ariadne" and "Pathway Studio" and older software versions, to find these publications. This list of PUBLICATIONS is provided.
Benefits of MedScan
- Create targeted databases for specific topics, organisms, diseases etc
- Highlight proteins, chemicals, cell processes, diseases, and other terms in the text
- Get instant informational snapshots or digests of your documents
- Detect entity co-occurrences in documents or sentences
- Set up automatic data extraction pipelines for regularly updating information content