alkyne search gives two hits for one alkyne

I'm search for the number of alkynes or alkynyl anions in a compound, using a substructure search with these parameters:

JChemCompare.jsp: setStereoSearch = 0

JChemCompare.jsp: considerDoubleBondStereoMatching = false

JChemCompare.jsp: considerOddCumuleneStereoMatching = true

JChemCompare.jsp: considerAxialStereoMatching = true

JChemCompare.jsp: considerSynAntiStereoMatching = true

JChemCompare.jsp: stereoMatchingModel = 0

JChemCompare.jsp: chargeType = 1

JChemCompare.jsp: radicalType = 1

JChemCompare.jsp: isotopeType = 1

JChemCompare.jsp: valenceType = true

JChemCompare.jsp: bondVagueness = 0

JChemCompare.jsp: orderSensitive = false

JChemCompare.jsp: listAllMatches = true

JChemCompare.jsp: target = [#6]C#C[#6]

JChemCompare.jsp: query = [$(C#C),$(C#[CX1-])]

JChemCompare.jsp: searchResult = true

JChem 5.9 is returning two hits:

The atom indices of the matches are: 
     [1]
     [2]
Each number represents an atom in the target; its position indicates the atom in the query to which it corresponds.

Why does JChem return two hits, and why does each consist of only a single atom?  If my SMARTS string is incorrect, can you suggest a better one?  I know I can set charge matching to default, but I'd rather have the SMARTS string specify the charges that may match.

Hi,

Your query,

[$(C#C),$(C#[CX1-])]

searches a single sp carbon atom, since one [...] unit represents only one atom according to SMARTS notation, specified with the chemical envinronment, $(C#C),$(C#[CX1-]).

 

Please find two solutions below that may fit to your needs:

1/ A query,

[$(C#C),$(C#[CX1-])]#[$(C#C),$(C#[CX1-])]

would match two sp carbons connected by a triple bond.

 

2/ Or this can be a very simple solution:

C#C

where two carbon is connected by a triple bond

 

Both query will give the same results:

Match 1:[    2,   3 ]

 

Best Regards,

Miklos

Your solution does not work for me.  I am trying to find a definition that will capture both alkynes and alkynyl anions. The query [$(C#C),$(C#[CX1-])]#[$(C#C),$(C#[CX1-])] does not match to CC#[C-].

Hi,

this behaviour can be due to the following option

--charge:d/e/i                Charge matching mode: d-default, e-exact, i-ignore

If you set the charge matching mode from "exact" to "default", the following example would work for you:

$ jcsearch -t:s --allHits --charge:d -q '[#6;X2+0,X1-1]' 'CC#[C-]'
    Query has 2 matches:
        Match 1:[    2 ]
        Match 2:[    3 ]
CC#[C-]

I hope this helps.

Best Regards,

Miklos

Yes, I know, but I was trying to devise a way to do the match without changing to default charge matching.  All my other functional group definitions use exact charge; this one is the only exception.  A while back I suggested that there be a way of specifying "exactly charge 0 or -1" in SMARTS, but y'all didn't seem very enthusiastic about doing so.  Anyway, I want a SMARTS expression that means C#[C charged 0 or -1] so that I can use exact charge matching in the MolSearch.  

Hi,

The exact charge matching mode would only match if the charge is explicitly defined on the molecule.

Please find attached another solution using R group query.

Doing this with SMARTS, it can only be viable if the charge matching mode is default.

Best Regards,

Miklos