Pages

Saturday, 26 March 2016

New research report tool

Quick post:

I've recently found a new tool which may be of use who like reading research papers. I was introduced to this at the rare disease conference I went to on rare disease day. (More on that in another post).

The tool is a website which references the pubmed database, and allows you to search for papers on any (rare disease) topic. HSP is included:

https://rareomics.healx.io/disease/spastic-paraplegia-hereditary

This then gives you the most recent research papers, from 2000, and these are indexed. There are 2 powerful tools included, the ability to search within the results by other search terms, and a set of trends/graphs.


(update 5th April:)

The trends and statistics part shows graphs of;

  • The number of papers per year for HSP for the last 10 years - which shows a peak in 2014, following a similar peak in 2009.
  • The top 5 genes mentioned in papers in the last 10 years, which shows that the five most common genes are SPG7, HSP90B2P, SPAST, SPG11 and ATL1. There is no particular trend in these, all following lines that are roughly horizontal.
  • The top 5 authors of papers in the last 10 years, who are Stevanin G, Santorelli FM, Blackstone C, Durr A, Brice A. Similarly to the genes, this shows roughly horizontal lines indicating that there are no particular trends.
I draw from this that the HSP research is reasonably well established, and that the dominant topics and authors are well established. This data looks similar to my own data analysis, which shouldn't be too surprising as it uses the same data source! It is interesting to note that this tool draws 927 papers from 2000, whereas I have only a few more than this with papers going back over the second half of the 1900's.

The news feed tab can be used to look up specifics, so you could, for example look up papers about "treatments" which it returns 114 papers (today, out of 927 papers) which include the term "theraputics". These can then be filtered using the summaries on the left, for example on "muscle spasticity" which gives 7 papers. Alternatively you can look up particular treatments - baclofen (8 papers), pain (10 papers), depression (3 papers), Botulinum Toxins (5 papers), dalfampridine (1 paper). This works OK, but you have to get the right term - for example "botox" returns zero papers.

Another feature is that you can list the results by "impact", so you can see which papers are regarded as having the greatest impact by looking at the number of times a paper has been referenced by other papers and the impact of the journal in which the paper is published. 

This seems really well put together.

Note to self, I really need to update and re-present my analysis of this data!

No comments:

Post a Comment