Dragon computes 3D but not 1D Descriptors

[Emma]

Hi all,

I have been using E-Dragon for computing molecular descriptors for quite some time but recently have come across a funny problem. Based on one of the widely cited work by Jorissen etal (J. Chem. Inf. Model., 45 (3), 549 -561, 2005), I used the Jorisson/Gilson Data Set to calculate molecular descriptors in E-Dragon.

The input file is a sdf file where the first 125 compounds are inhibitors and rest are non-inhibitors.

Problem:E-dargon computes 1666 descriptors for the entire dataset.However , only in the case of inhibitors [first 125 compounds] it doesn't compute the following specific constitutional descriptors,
MW molecular weight
AMW average molecular weight
Sv sum of atomic van der Waals volumes (scaled on Carbon atom)
Se sum of atomic Sanderson electronegativities (scaled on Carbon atom)
Sp sum of atomic polarizabilities (scaled on Carbon atom)
Ss sum of Kier-Hall electrotopological states
Mv mean atomic van der Waals volume (scaled on Carbon atom)
Me mean atomic Sanderson electronegativity (scaled on Carbon atom)
Mp mean atomic polarizability (scaled on Carbon atom)
Ms mean electrotopological state

These are the most simplest descriptors which need only the molecular composition of the compound for calculation.Therefore, while E-Dragon calculates the rest of 3D descriptors I haven't found a logical reason why it should not compute the above given 1D descriptors.

Hence , if anyone has any suggestion or ideas please let me know. Any comments are highly appreciated.

Thanks,
Emma

[Andrea Mauri]

Dear Emma,
I am one of the developer of edragon.
Could you please send me a link where I would be able to download the Jorissen/Gilson dataset?
I will check it and post a reply.
Andrea

[Emma]

Dear Andrea
I have been able to figure out the error. In the output file Dragon calculated the specific 1D descriptors ,however, it used to create spaces after the molecule name.As a result i concluded that these descriptors was not being calculated but later on realized that the columns were only shifted.
The reason for this error was simply that in the input sdf file there were spaces after the molecule name.

Also,Igor Tetko has already corrected this problem and now I get the correct output file.

Thanks,
Emma

[Andrea Mauri]

Dear Emma,
once I downloaded the Jorissen-Gilson dataset I encountered the same problem but I was not sure you used exactly the same dataset.
Anyway I am happy to read that Igor already fixed it.
Andrea