Legio stereo bug

[Lars Kolster Petersen]

Dear GGA software

 

I have been investigating the use of Legio as I have an interest in multi-building block combinatorial libraries. I have been puzzled to find out that Legio handles enantiomer building blocks poorly. In a very simple test (see attached .sdf) I loaded two mirror images of phenylalanine (don’t mind the N-I modification – it is for other use) and subjected them to a rxn file that simply transforms the carboxylic acid into the corresponding benzylamide. In the product sheet only one is observed even though both enantiomers should yield distinct products.

I have two questions:

1)       Is it possible to make Legio handle enantiomeric building blocks correctly.

2)       Is it possible to have Legio save only SMILES string and reactant code into a text file rather than as an .sdf? For larger libraries (10.000+) the output files get rather larger…

 

Also, my attempts on initiating Indigo have not been fruitfull. However, I am not that used to working with Java…

 

Best regards

 

Lars Kolster Petersen, PhD
Vipergen

Gammel Kongevej 23A, 1st floor

DK-1610 Copenhagen V
Denmark
+45 25236680

l...@vipergen.com

web www.vipergen.com

-----------------------------------------
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify Lars Kolster Petersen.

 

[Mikhail Rybalkin]

Hello Lars Kolster Petersen,

We have paid a lot effort to handle stereochemistry in Legio correctly.� You can look at the
example on your webpage: http://ggasoftware.com/opensource/indigo/concepts/combichem

There is no chiral flag for your molecules in the file Phe+stereo+test.sdf. According to molfile
specification if there is no chiral flag then the structure represents a mixture (this structure
and mirrored one). You can add this flag in your editor for the whole structure, or mark your
stereocenters as absolute. I have attached both corrected versions:
Phe+stereo+test.sdf-chiral.sdf and Phe+stereo+test-abs.sdf.
Legio produces two different products with them.

2)������ Is it possible to have Legio save only SMILES string and reactant code into a text file rather than as an .sdf? For larger libraries (10.000+) the output files get rather larger�

It is not possible to do this with current version, but we will add this in the next version.

Also, my attempts on initiating Indigo have not been fruitfull. However, I am not that used to working with Java�

Could you explain what kind of problem you had? I can provide you the small code snippet
for using reaction product enumerator in Java if necessary.�
�
Thank you for the interest in Legio!

Best regards,
Mikhail

On 10/13/2011 00:10, Lars Kolster Petersen wrote:

Dear GGA software

�

I have been investigating the use of Legio as I have an interest in multi-building block combinatorial libraries. I have been puzzled to find out that Legio handles enantiomer building blocks poorly. In a very simple test (see attached .sdf) I loaded two mirror images of phenylalanine (don�t mind the N-I modification � it is for other use) and subjected them to a rxn file that simply transforms the carboxylic acid into the corresponding benzylamide. In the product sheet only one is observed even though both enantiomers should yield distinct products.

I have two questions:

1)������ Is it possible to make Legio handle enantiomeric building blocks correctly.

2)������ Is it possible to have Legio save only SMILES string and reactant code into a text file rather than as an .sdf? For larger libraries (10.000+) the output files get rather larger�

�

Also, my attempts on initiating Indigo have not been fruitfull. However, I am not that used to working with Java�

�

Best regards

�

Lars Kolster Petersen, PhD
Vipergen

Gammel Kongevej 23A, 1st floor

DK-1610 Copenhagen V
Denmark
+45 25236680

l...@vipergen.com

web www.vipergen.com

-----------------------------------------
This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify Lars Kolster Petersen.

�