From npsc;'s comments in the blog I distill that we explicitly need to share a ligand with scientists to get it noticed. Is that correct ?1
If so, why are ligands not automatically shared when they are present in a ligand search ?! I rarely share a ligand because I assumed there would be a mechanism in place to forward anything that increases score and fulfills the objectives.
Please enlighten us.
In order for a ligand to be considered, the "solution" containing it has to be sent to the server. This happens for shared solutions, but also for your personal top scoring structures, as well as occasionally for intermediate work while you're playing online. You don't have to explicitly share with scientists to have your ligand be considered, but if you find an interesting ligand & binding position it might be worth sharing-with-scientists just to make sure it gets considered (especially if it's not a top-scoring one) and that the automatic mechanism doesn't accidentally skip it.
The compound library search only provides the ligand identity to the server, not the ligand bound to the protein. For final results, we're looking for those ligand+protein structures, so we don't really consider the search submissions.
In most cases that automatic mechanism should capture most of what's interesting. It will typically capture the top scoring solutions and those ligands you're spending the most time on. For most players, that should be sufficient. The share-with-scientist mechanisms is only an additional facility to allow those who are interested a way of making sure particular ligands are included in the analysis.