OpenBABELでSMILESやInChIから初期座標自動生成するのは結構良い

OpenBABELでSMILESやInChIから初期座標自動生成するのは結構良い

以前CBI学会でPubChemQCプロジェクトで、初期座標はコレでエエのかと言われましたが
そのときは、分子量が200以下ならほとんど配座は関係ないですとかとりあえずごまかしました。

で、実際はどうなってるんだろう？と思ってドキュメント調べたら結構真面目にやってるらしい。
以下引用

gen3d

To illustrate how some of the above methods might be used in practice, consider the gen3d operation. This operation (invoked using --gen3d at the commandline) generates 3D structures for 0D or 2D structures using the following series of steps, all of which have been described above:

Use the OBBuilder to create a 3D structure using rules and ring templates
Do 250 steps of a steepest descent geometry optimization with the MMFF94 forcefield
Do 200 iterations of a Weighted Rotor conformational search (optimizing each conformer with 25 steps of a steepest descent)
Do 250 steps of a conjugate gradient geometry optimization

Taken together, all of these steps ensure that the generated structure is likely to be the global minimum energy conformer. However, for many applications where 100s if not 1000s of molecules need to be processed, gen3d is rather slow. A future version of Open Babel will provide options for slow/medium/fast 3D structure generation which will involve different compromises between speed and finding the global energy minimum.


幾つかコンフォーマー作って比較するというのは必要ないかもな。