Review of Medminer



Title:  Medminer

URL:http://discover.nci.nih.gov/textmining/filters.html

Date Reviewed:May 20th 2002

Description: The MedMiner filters will extract and organize relevant sentences in the literature based on a gene, gene-gene or gene-drug query. This tool combines the GeneCards and PubMed search engines with user input and automated server-side scripts in an integrated text filtering system.

GeneCards is a database of human genes, their products and their involvement in diseases. It offers concise information about the functions of all human genes.

Strengths: The program is designed that excisting databases will be used more efficiently. For it's role in for example micro-array analysis one can ask interesting questions as:

Question: What genes on the chip are related to apoptosis? It uses genecrads

Question: How do I find out more about the gene Caspase 3 with GenBank Accession number U13737? It uses pubmed.

It gives you a clear example how to use the filter systems and how to upload your won data, what to do with problems Etc the tutorial therefore is very helpful.  (see below)

 Weaknesses:

I think it is weak that one can only look for interaction between either two genes at the same time and or drugs. Maybe a whole pathway and or related genes would be more helpful. However one has to put in so many keywords it becomes almost impossible to see what the filter system is actaully doing.

One first has to enter to genes, then some keywords and then you get your list. But what has to do with what? Howver it gives directly a list of genes that are also expressed on you chip so that is a nice feature.

Comprehensiveness:

For the tools is has it is relatively handy, it gives you a list with frequently used keywords in the various fields, however my guess is that if one would like to know about certain genes a quick pubmed search only would be the first thing people would try.

Timeliness: The actual search and filtering took me a lot of time. Maybe the filtering is too time consuming, I looked up two genes p53 and LMp1 involved in viral oncogenesis. Then I had to add two more keywords and I got 1835 hits. However the program give you an extra set of keywords (related) which have decreased number of hits.

Ease of Use: It is very easy to use, but the link between the array data and the filter system remains somewhat obscure. Yes if you find two genes up-regulated on your chip it is nice to find papers about the correlation but that is also possible with a normal pubmed search.

Responsiveness:

Similar or Related Sites: Pubmed and Genecards

Overall Evaluation: 1 to 5 stars 2*

Reviewer: D.M.Pegtel

Email:dirk.pegtel@tufts.edu