the answer to your first question depends on your situation, e.g.:
what is the origin of your data and how do you use them. For example, if the original data is in pdb format and you can also read/use them as pdb, then there is no reason to convert them into anything else.
what type of information must be represented ? One smiles string would be enough to represent the connectivity between atoms. Three dimensional coordinates require different formats and experimental data still more ?
How many entries do you handle ? If you have >> 100 molecules you will want to make sure that you have no duplicates in your list, which could be achieved by a canonical smiles or an inchikey.
I'm not that familiar with pdb, it seems a common but rather complex and inflexible format to represent e.g. proteins, also including crystallographic data.
This may not be your first choice if you need to represent e.g. conformers of small molecules, but even then: if you have your data in pdb and tools to read pdb format you may well want to stick with it ?
I'm not sure how to understand you 2nd question: first, you definitely need to add hydrogens before generating 3D coordinates. Generating one conformer per molecule is straightforward with Babel, but I dont understand what you mean with "the best conformer for 50 molecules" ?
On Tue, Feb 28, 2017 at 5:11 AM, Mohammad Goodarzi <[hidden email]> wrote:
I have two questions,
1- I have my molecules as .pdb formats , is it ok? or should I convert them to another format? if so, should I add hydrogen too?
2- I generate many conformer for each molecules, is there a way to select the best conformer for lets say 50 molecules ? do you have any python or perl solution for it?