Patent structures as reference compounds

17 Oct
Published by cdsouthan

Subject 

Request for Help

The recent small molecule anti-malarial patents review by the MMV teams is a very useful  job (http://www.citeulike.org/user/cdsouthan/article/11477246) and  the effort raises some questions.
 
1)  It would obviously be valuble to expidite an open deposition of the 50 or so key structures (InChI, SMILES, IUPAC mol etc) linked to patent IDs (figshare would do fine in the first instance).  This could save major effort for the many interested parties in this comunity having to run  them through OSRA, sketching or otherwise consolidating patent mining efforts via chemicalize.org or SureChemOpen.  As the SciFinder stucture images are declared in the paper dropping out  their mol files should be OK. Maybe the authors have even collated the structures already.  If there are any other possible impediments on the journal side these can be negociated in the OSDD context.  I would guess anwhere between 50 -80% of the structures are in PubChem already.   
 
(e.g.  http://cdsouthan.blogspot.se/2012/09/mmv390048-promising-new-antimalaria...
 
The CIDs would thus be a useful  X-ref.  Some may also be in ChEMBL if the papers have appeared in the journals they extract but these patents are fairly recent  ( I may get round to folowing up  a few from the review in a future blog post).
 
2) There is also obvious value in getting  at least some of these into the "Malaria Box" collection, as refrence compounds and mechanistic probes, even via requests to the assignees (nothing to be lost by asking anyway)
 
3)  It would seem likely that most inventors and assignee organisations for antimalarial patents are now aware of  OSDD Malaria activities and this website.  There would thus be a major community benefit if , soon after the publication of any new patents (or older  ones that might have slipped through the MMV trawl) these were "surfaced" in some way.  Somewhere on this site would make sense but note that patent citations can be entered in CiteUlike and/or Mendely, tagged, and the applicants could write their own abstracts as well.
 
4)  Faster and more comprehensive surfacing of key active compounds from patents expands the opportunities for SAR extraction, in silico modelling and in vitro profilling
 
5) The same considerations for structure deposition apply to this 2007 -10 patent review  (which I cannot acess) http://www.citeulike.org/user/cdsouthan/article/11463021   if Gemma et al pick up this post (or someone forwards it please)
 
6)  Please comment here if there is a strong comunity interest in a) antmalarial patent mining for small-molecule SAR, model building,  and b)  the implementation of  panel-type standardised cross-screening  of selected structures (in // with the leads from the Sydney project obviously)  with result deposition into both ChEMBL-NTD and a PubChem Bioassay.   
 
7)  I assume the analogous process for papers is probably practiced already, e.g. by the more chummy teams swapping leads,  but the same principal of the sharing power applies here as well  and some papers may be slipping through the gaps.   The ideal case woul be a Mendely group where key structures from papers had public databases IDs and even eventulally found their way into  the Malaria Box
 
 
 
 

Comments

MatTodd's picture

Yes, all of what you suggest would be very useful. We need to be keeping a closer eye on patents for a) interest and b) relevance. The best way to do that is to, as you say, "surface" points of interest.

 

For the other stuff, if you recommend people should take action, one way to encourage that is to split into a work plan of microcomponents. What can we do to do this - e.g. to make the first point happen?

 

And - sorry - are we still looking for a copy of PMID:20738242?