Can JCHEM recognise imposed proton in a substructure search

I would like JCHEM search to take into account imposed H and thus retrieve only structures with these imposed H.





For instance I would like the following substruture search [H]C1=C([H])N=C(N)N1 not retrieve as many structures as the following substructure search NC1=NC=CN1 will do





Thanks for your help

POJ2 wrote:

I would like JCHEM search to take into account imposed H and thus retrieve only structures with these imposed H. For instance I would like the following substruture search [H]C1=C([H])N=C(N)N1 not retrieve as many structures as the following substructure search NC1=NC=CN1 will do

JChem considers explicit H atoms in the query, so only retrieves structures with explicit or implied Hydrogens at the given position.


(However, explicitly drawn or implied H atoms cannot be distinguished in the target .)





For example, the molecule NC1=NC2=C(N1)C=CC=C2 is retrieved by your second query but not the first.





For more examples, please see the relevant section of the JChem query guide:


http://www.jchem.com/doc/user/Query.html#explH





Please do not hesitate to ask if you need any further information.





All the best,





Szabolcs

ok


but why does the substructure search on [H]OC1=C([H])C=C(Cl)C=C1[H]


retrieve the following structure : COC1=C(OC)C=C(CO)C(Cl)=C1 ?





Thanks for helping

POJ2 wrote:

ok but why does the substructure search on [H]OC1=C([H])C=C(Cl)C=C1[H] retrieve the following structure : COC1=C(OC)C=C(CO)C(Cl)=C1 ?

That's strange, I cannot reproduce:

Code:

$ jcsearch -q "[H]OC1=C([H])C=C(Cl)C=C1[H]" "COC1=C(OC)C=C(CO)C(Cl)=C1"

(No output means no hit.)





If I remove two H-s from the query, there is a hit, which is correct:

Code:

$ jcsearch --allHits -q "OC1=C([H])C=C(Cl)C=C1" "COC1=C(OC)C=C(CO)C(Cl)=C1"     Query has 1 match:         Match 1:[    5,   4,   7,  -7,   8,  11,  12,  13,   3 ] COC1=C(OC)C=C(CO)C(Cl)=C1

In what context did you experience the wrong behaviour? (MolSearch API, database search, jsp interface etc.) Which version of JChem are you using?

sorry, for the trouble; I found the cause of my substructure retrieving bad structures : I used to turn explicit H to implicit ones (using hydrogenize(false) method) before lauching the search (which I do for compound registration)

No problem. I'm glad you found the bug.