Sign in or Join FriendFeed
FriendFeed is the easiest way to share online. Learn more »
Matthew Todd
Upload chemical procedures to a single source or just put them anywhere and have people/spiders find them? I've posted a sample synthetic procedure on the Synaptic Leap site ( and uploaded similar data to Chemspider ( Preferences?
Preferences? Aggregated/encyclopedic source, or distributed? Are the data not on chemspider found/retrieved effectively? - Matthew Todd
Distribute and duplicate - maximize the chances of someone finding your stuff and actually using it - Jean-Claude Bradley
Right, but the question is, would people find my original synaptic leap description? Answer - only those people who know to search with InChi. And that's just for the product. Nobody searching for 'Boc-protection of amines' would find it. So is the original post any use? - Matthew Todd
And with distributed comments ( don’t think it hurts to put it anywhere that you trust as long as the aggregation works. Is it more convenient to you to host locally on some other service? It’s all early days anyway - none of this works as well as it might so exploring the options is the way forward in my view. Ideally Chemspider/ChemBlogSpace/Others would aggregate all mentions via whatever identifiers and everything gets nicely indexed at multiple services - we’re some way away from that yet. - Cameron Neylon
But I would argue that the synaptic leap version would be better if you used the embed feature from Chemspider for the molecules in the scheme - then CS knows that you're connecting back and can check whether to aggregate and markup back. What might also need to be exposed in a little bit of RDF/Microformats saying "this is a chemical synthesis" so CS can make an informed decision as to what to do with it - Cameron Neylon
We can't embed reactions, yet, right? - Matthew Todd
Mat as long as the Synaptic Leak allows indexing by search engines people will find your stuff. For example searching for "boc-glycine solubility" on Google gives our lab notebook on the second result and our database on the 9th of the first page - Jean-Claude Bradley
Mat as for reactions, Peter Murray-Rust Egon Willighagen, Cameron Neylon and I have experimented with representing our Ugi reaction using CMLReact but nothing systematic yet : - Jean-Claude Bradley
I'm not so sure - imagine the web is full of synthetic procedures (what a vision) - current searches for molecules would only be satisfactory if all those pages contained e.g. InChIs - boc-glycine is usefully precise, but many other names are too complex. There are no InChIs for reactions as yet. - Matthew Todd
...and I don't know of any organic chemists who know what InChIs are, and I know a fair few... So are we currently relying on people complying with putting in InChIs on their own pages? To me this is an advantage of Chemspider, where the data are presented in a pre-arranged way. - Matthew Todd
Matthew, most commercial and opensource offerings have a "Create InChI" functionailty now... it's turn-key technology, just like "Create IUPAC name"... - Egon Willighagen
InChI: unique identifier for (small, organic) molecules - Egon Willighagen
Sort of like and email address, but then for molecules... - Egon Willighagen
That's a simple message, I'd say. If they ask why they should use it, you can say that they kind find a lot of *experimental* data on the internet by using the InChI as query on Google. - Egon Willighagen
Just thinking aloud here... actually... not really replying to something in particular I think... I wish I had the time to set up - Egon Willighagen
Mat, ideally if you used e.g. the chemspider embed function (and you can therefore find the molecule by name or structure) then that would expose the InChi on the web page where you embed it as part of the javascript. Then the experimental chemist wouldn't even have to worry about it. It would just be there automatically whenever you linked. - Cameron Neylon
Mat we also tag our experiments with InChIs and InChIKeys but the reality is that the vast majority of Google searches that find our pages are via common names - it isn't an either/or situation - just use as many ways as you can with a reasonable effort - Jean-Claude Bradley
Egon - You're preaching to the converted - I think they're great, nobody else sees the point... Large databases such as SciFinder means never having to worry about how it's done. - Matthew Todd
Cameron - very interesting that the embed feature includes the InChI. That's useful. Again, it will take a lot for the average organic chemist to include this in a post, particularly if there are multiple structures in a reaction scheme/mechanism. - Matthew Todd