Open cdsouthan opened 8 years ago
Hi @cdsouthan. Thanks for this suggestion. I discussed something like this a year or so ago with some people about the compounds that did make it into the MMV boxes: how about we do something to make sure that the malaria box or pathogen box compounds have readily available, open synthetic routes and/or are always physically available in some way? I had heard that people who had discovered interesting things with the MBox compounds wanted to follow up those findings, but then were stuck on how to get more. I don’t know to what extent that was a problem - perhaps a biology lab with no available chem collaborators who wanted compounds that had arisen from libraries and for which no bulk material was available? Perhaps CRO resynthesis was expensive if there was no published precedent. I’m sure ways were found in the cases of particularly promising compounds.
A paper is I think forthcoming ( @PaulWillisMMV ? ) on the experience on the MBox, but as I understand it people in the end took a variety of approaches to source more compounds, ranging from purchase through to local resynthesis. Obviously in OSM Series 1 we used resynthesis. These solutions are pretty generic and would work for good-looking compounds that didn’t make the cut for inclusion in the Pathogen Box.
For the Pathogen Box compounds themselves I’m not sure of the current/proposed approach. Perhaps the CROs involved will be willing to share lab notebooks, though that still leaves the pre-existing commercial compounds. My sense is that there should be a site somewhere (we could easily host, and have offered to do so) in which each compound has a chemical recipe, and people who attempt to reproduce that work can post their own suggestions/spectra/data. A community-driven compendium of resynthesis. Clearly this would work too for the compounds that did not make it, and is at the heart of what OSM does.
The other option is what you suggest - having some compounds of interest be commercially available. There is merit to this even if the data are open since in many cases people may want to buy for convenience/speed or because there is no local synthesis resource. If there’s a market, then why not? It is likely this is already being considered by suppliers to the Pathogen Box.
As far as OSM is concerned we can certainly nominate exemplar compounds from each series, plus ample synthetic precedence. SAR controls are easy if you want negatives :) Could be a nice way to promote continued interest in compounds we may have put on back burners, and yes to help promote their broader evaluation.
Is this what you mean? OSM compounds, and those that didn’t make it, being offered commercially even when there are open (source) routes to their synthesis available?
Hi,
The structures of the Pathogen Box compounds have been available for some time, as announced on the Pathogen Box website http://www.pathogenbox.org/newsroom/news/structures-400-compounds-pathogen-box-now-available
The Pathogen Box team plans to make the experimental procedures available for all the synthesized compounds and to list the commercially available compounds. It will take some time to compile this information - ~ 900 compounds were screened to arrive at the 400 included. If groups need info on synthetic routes/availability of a particular compound sooner, please contact us and we can provide the info. We aim to have material in stock available to supply to groups to allow them to perform follow up studies on particular hits
MMV synthesized or purchase additional stocks of Malaria Box compounds and did supply samples to many groups who contacted us. We therefore encourage groups to contact us once they have screening hits from Malaria or Pathogen Box compounds so we can discuss ways we can help groups progress their hits
Paul
Paul Willis| Medicines for Malaria Venture Director, Drug Discovery Tel: +41 22 555 03 42 Mobile: +41 79 102 78 74 Email: Willisp@mmv.orgmailto:Willisp@mmv.org
Defeating Malaria Together | www.mmv.orghttp://www.mmv.org/
From: Mat Todd [mailto:notifications@github.com] Sent: Wednesday, 27 January 2016 11:57 AM To: OpenSourceMalaria/OSM_To_Do_List Cc: Paul Willis Subject: Re: [OSM_To_Do_List] Vendor interest (#369)
Hi @cdsouthanhttps://github.com/cdsouthan. Thanks for this suggestion. I discussed something like this a year or so ago with some people about the compounds that did make it into the MMV boxes: how about we do something to make sure that the malaria box or pathogen box compounds have readily available, open synthetic routes and/or are always physically available in some way? I had heard that people who had discovered interesting things with the MBox compounds wanted to follow up those findings, but then were stuck on how to get more. I don’t know to what extent that was a problem - perhaps a biology lab with no available chem collaborators who wanted compounds that had arisen from libraries and for which no bulk material was available? Perhaps CRO resynthesis was expensive if there was no published precedent. I’m sure ways were found in the cases of particularly promising compounds.
A paper is I think forthcoming ( @PaulWillisMMVhttps://github.com/PaulWillisMMV ? ) on the experience on the MBox, but as I understand it people in the end took a variety of approaches to source more compounds, ranging from purchase through to local resynthesis. Obviously in OSM Series 1 we used resynthesis. These solutions are pretty generic and would work for good-looking compounds that didn’t make the cut for inclusion in the Pathogen Box.
For the Pathogen Box compounds themselves I’m not sure of the current/proposed approach. Perhaps the CROs involved will be willing to share lab notebooks, though that still leaves the pre-existing commercial compounds. My sense is that there should be a site somewhere (we could easily host, and have offered to do so) in which each compound has a chemical recipe, and people who attempt to reproduce that work can post their own suggestions/spectra/data. A community-driven compendium of resynthesis. Clearly this would work too for the compounds that did not make it, and is at the heart of what OSM does.
The other option is what you suggest - having some compounds of interest be commercially available. There is merit to this even if the data are open since in many cases people may want to buy for convenience/speed or because there is no local synthesis resource. If there’s a market, then why not? It is likely this is already being considered by suppliers to the Pathogen Box.
As far as OSM is concerned we can certainly nominate exemplar compounds from each series, plus ample synthetic precedence. SAR controls are easy if you want negatives :) Could be a nice way to promote continued interest in compounds we may have put on back burners, and yes to help promote their broader evaluation.
Is this what you mean? OSM compounds, and those that didn’t make it, being offered commercially even when there are open (source) routes to their synthesis available?
— Reply to this email directly or view it on GitHubhttps://github.com/OpenSourceMalaria/OSM_To_Do_List/issues/369#issuecomment-175556661.
This email and any attachments may contain confidential information. If you are not the intended recipient, please delete the email and any attachments and notify the sender immediately.
@madgpap It would be useful if all these compounds were registered in ChEMBL.
Thanks for useful reply @mattodd. OK, so we converge on the main ways NTD researches can get hold of other folks leads via the boxes, open synthetic schema and vendors. These can all be useful routes under different circumstances for different teams.My suggestion is a) you come up with three pairs from the major series (but not in the box) b) they dont have to be the most potent leads but if there is say 10-fold potency difference (idealy from two independant assay results) and c) where the negative control (decoy or dud in QSAR modelling parlance) was not too synthetically divegent from the active (i.e. semi // synthesis vi mostly common steps for the pair would lower the cost). Note also you could point to the open synthetic data anyway so some labs might still be prepared to DIY to get such useful cross-screening pairs, even if the appeared on vendor offerings as well. JFTR my vendor contact is not inclined to stealth mode but they do need to cost out carefuly to asses feasibility and commercial consideration. Note also the proposal is an unsual one iin that, like most vendors, only development candidates typically make in to the catalogue.
I'm on it.
Sent from my giPhone
On 27 Jan 2016, at 14:26, Chris Swain notifications@github.com wrote:
@madgpap It would be useful if all these compounds were registered in ChEMBL.
— Reply to this email directly or view it on GitHub.
@PaulWillisMMV Thanks for the pointer to the PathBox data sheet. However, I can only map 358 of those 400 SMILES into PubChem.
http://www.ncbi.nlm.nih.gov/sites/myncbi/christopher.southan.1/collections/49528439/public/
I noted you stripped the salts back but I don't think thats the problem. The PubChem uploader accepts the unmatched ones as valid SMILES so it looks like they are just not in there as exact matches. I cant check the missing ones individually just now but it would be really usefull if you could assign CIDs to all of them somehow, especially since they should all be established (published also?) structures.
You can see the missmatches in this file PathBox_400_SMILES_vs_CIDs.xlsx
n.b.1 if @madgpap gets the 400 as a set into ChEMBL21 they can eventually come into PubChem that way n.b.2 243 of the 358 have vendor matches from which box users could replenish from if they needed
Consequent to a colloquial vendor contact at a recent meeting (who I need not identify just yet) we discussed the idea of (them) offering selected OSM leads for sale and/or other leads from that are likely to not have made it into the MMV Pathogen box (when we finaly get to see the structures that did make it)
http://cdsouthan.blogspot.se/2014/06/getting-into-box-with-some-recent.html
While a commercial exersise by definition, it would still be a way of expediting open data sharing. This would be particularly so since lead swapping for antimalarial teams to cross-test probaly does not happen as much as it could and open groups obviously sythesize at a relatively small scale that make this difficult anyway. This could also facilitate molecular target cross screening for mechanism of action deconvolution.
If this idea has any merit would anyone from OSM care to pick out maybe one lead from each series that could be canidates for this? (I could have a go but dont know the sets so well). NOBA, choosing chemotype-matched SAR controls could also be useful (i.e usual principal of chemistry as similar as possible but with activitues as far apart as possible - at least an order of magnitude)