tutorial for MX

[Sean]

I am new to MX but I see it is tremendously useful. Can anyone kindly
point to any online tutorials?

Regards,

Sean

[rapodaca]

Hello Sean,

There are several tutorials, that could be compiled into a single
guide. So far that hasn't happened yet.

The home page is probably your best bet for getting started quickly:

http://code.google.com/p/mx-java/

At the bottom are links to tutorials written as blog articles.

Let me know if you're looking for something more specific.

Cheers,
Rich

[Sean]

Thanks Rich. I have run into a problem with using the
SMILESReader.read() method. I was trying to read some known SMILES
string e.g. CCCCCCCCCCCCCC[NH2+]CCCC[NH+](CCCCCCCCCCCCCC)CC(C[NH3+])O.
[Cl-].[Cl-].[Cl-] and I am getting "Invalid SMILES character in
"CCCCCCCCCCCCCCC[NH2+]CCCC[NH+](CCCCCCCCCCCCCC)CC(C[NH3+])O.[Cl-].
[Cl-].[Cl-]". Any advice?

This is line of Java code in my test file:

Molecule UC0254 = SMILESReader.read("CCCCCCCCCCCCCC[NH2+]CCCC
[NH+](CCCCCCCCCCCCCC)CC(C[NH3+])O.[Cl-].[Cl-].[Cl-]]");

Thanks a lot.

Sean

> > Sean- Hide quoted text -
>
> - Show quoted text -

[rapodaca]

Hello Sean,

Unfortunately, the SMILES parser implements only a subset of SMILES at
this point. The error output should really be more specific - thanks
for bringing it up. My guess is that the one of the "[", "+", "-", and
"." characters is responsible.

To get an idea of the kind of SMILES currently supported by MX, check
out the test package:

http://github.com/rapodaca/mx/blob/039a8cd74ca4fc6cf9bedec01325cbd11c110720/src/com/metamolecular/mx/test/SMILESReaderTest.java

SMILESReader is set up to throw an exception on encountering any
character that it doesn't understand, even if that character is part
of the 'spec'. As SMILES features are added to MX, these errors will
(hopefully) go away.

In short, reading and writing SMILES is a work in progress. Completing
it is currently a low priority over here, but I'd be happy for someone
such as yourself to look into moving the implementation forward.

Cheers,
Rich

[Sean]

Thanks Rich. I am interested using the SMILESReader to read in the
string and do substructure search using the DefaultMapper. Could you
please explain how involved it is to fully implement the spec to
enable the search? What is the mechanism of sub structure search and
how is that related to the SMILESReader? If it sounds like an
achievable goal, I or one of my colleagues might be able to help.

Sean

On Mar 11, 6:05 pm, rapodaca <rich.apod...@gmail.com> wrote:
> Hello Sean,
>

> Unfortunately, the SMILES parser implements only a subset of SMILES at
> this point. The error output should really be more specific - thanks
> for bringing it up. My guess is that the one of the "[", "+", "-", and
> "." characters is responsible.
>
> To get an idea of the kind of SMILES currently supported by MX, check
> out the test package:
>

> http://github.com/rapodaca/mx/blob/039a8cd74ca4fc6cf9bedec01325cbd11c...

> > > - Show quoted text -- Hide quoted text -

[rapodaca]

Hello Sean,

A SMILES guide can be found here:

http://www.daylight.com/dayhtml/doc/theory/theory.smiles.html

There's also the OpenSMILES project:

http://www.opensmiles.org/

Regarding the mechanism of substructure search, it is based on the VF
algorithm:

http://depth-first.com/articles/2008/11/17/substructure-search-from-scratch-in-java-part-1-the-atom-mapping-problem

DefaultMapper can be used with Molecules generated in several ways -
for example from molfiles. The mapping package is independent of the
SMILES package. As long as you can access a valid Molecule, you can do
a match.

Cheers,
Rich