Removing Alkyl Groups

Hi,

This is my first post to the forums, so thanks in advance for all of your help.

I have two databases of molecules stored as SMILES, one with about 1000 molecules and the other with several million.  I would like to automate a process that removes alkyl tails from the molecules and generates the new SMILES for the molecules.  Is it possible to do that?

The database is coded in python with a Django interface, I don't know if that's relevant.

Thanks

Laszlo

Hi,

Could you post a few input and expected output examples here? For example what do you expect for these inputs?

CCCCCCCCC
CCCF
CCCCC(C)CC
CCCCCCCCC1CC(CCC(C)CC)CC(CCC)C1CCCCC

Zsolt

 

Hi Zsolt,

 

I suppose I should clarify that the molecules in my database largely consist of conjugated systems/rings with alkyl tails (for example:

 CCCCCCC1=CC2=CC3=CC=C(C=C3C=C2C=C1)C1=CC2=CC3=C(C=C(CCCCCC)C=C3)C=C2C=C1

or

CCCCCCCCOC(=O)C1=CC2=CSC(C3=CC4=C(OCC(CC)CCCC)C5=C(C=CS5)C(OCC(CC)CCCC)=C4S3)=C2S1

which would become: 

C1=CC2=CC3=CC=C(C=C3C=C2C=C1)C1=CC2=CC3=C(C=CC=C3)C=C2C=C1

and

COC(=O)C1=CC2=CSC(C3=CC4=C(OC)C5=C(C=CS5)C(OC)=C4S3)=C2S1

The molecules you gave are all alkyl.  For numbers 1, 3, and 4, I guess I wouldn't want to keep anything, and for 2  I would keep CF.

Thanks for your help!

Laszlo

I would try to remove terminal CH3 atoms with Standardizer using a quasy loop. This is the transformation scheme:

[CH3:1][#6:2]>>[H:3][#6:2]

 

Test with your second input:

$ standardize -c '[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:
1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..
[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6
:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:
3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]
>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][
#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH
3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]
..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][
#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[
H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:
2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1
][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[
CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:
2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3
][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>
>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#
6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]..[CH3:1][#6:2]>>[H:3][#6:2]' 'CCC
CCCCCOC(=O)C1=CC2=CSC(C3=CC4=C(OCC(CC)CCCC)C5=C(C=CS5)C(OCC(CC)CCCC)=C4S3)=C2S1
'
COC(=O)C1=CC2=CSC(C3=CC4=C(OC)C5=C(C=CS5)C(OC)=C4S3)=C2S1

 

In this test, I apply the same carbon terminal removal 50 times, so it removes alkyls up to 50 atoms. See more details here.

Hi,

 

I'm a little confused where to input the commands into standarizer.  When I start it up and give Standardizer the input file, it then gives me a list of things to do (such as aromatize, remove explicit hydrogens, etc.).  Where can I enter the command you gave me?

 

Sorry for my confusion.

 

Thanks,

Laszlo

What I sent you is a command for a console.

If you prefer the graphical user interface, you just need to select the the "Transform" action in the available actions list on the left, and add it to the right list. 

Then select that transform on the right pasting this transformation in its scheme editor:

{code}

[CH3:1][#6:2]>>[H:3][#6:2]

{code}

Now you have a transform to convert methyls to hydrogens. If you add the same transform 30 times to your action list, it will remove alkyl chains up to 30 carbon atoms.

To save some time for you, I have actually created the xml configuration, you can load it on that configuration page and run directly. Please find it attached.

Hi Gyuri,

 

It works great.  Thanks for your help!

 

Best,

Laszlo