Hello,
I am working with MOE (Molecular Operating Environment) and I need to generate a database of conformers for a .sdf database of more than 20 000 molecules for later pharmacophore screening. However, it has got stuck already after 500 molecules. Does anyone have a similar experience or any idea what could be a problem please?
Surprising, as MOE is usually quite good in generating conformers from libraries.
Try some of the usual checks...
The molecule: Can you perform a conformer search on the exact molecule(s) where the program gets stuck? Just to see if your issue is not purely linked to your compound?
If you suspect this is a memory / size problem: did you try to "chunk" your input file in smaller files?
Just to add to Sandro's answer - Are you performing those calculations on a laptop (as they are not very efficient in handling memory intensive tasks [especially the thinner ones]). You could also try similar calculations on a different computer (desktop/laptop) to rule out your computer related issues.
Also to see is whether the speed of ligand processing quicker for earlier 499 cpds or slower, it could just be that you have high rotatable bonds which could make everything slower in first place.
Thank you very much for your suggestions and sorry for my late reply! We have already found the solution, we launch a script for this work and everything is going good now.
Although it seems that you have already solved your problem, I'd like to add to this as well. I came across this same problem where MOE kept getting stuck during a large run, which became incredibly frustrating when working with large libraries. After ruling out file size and computer issues, It turns out that it was getting stuck at certain compounds where the MOE-evaluated energy (E) was incredibly high after minimization (I used the MMFF94x forcefield). To bypass this, I simply reordered the compounds according to this defined E value so the conformer search would get stuck where the calculated energies were too high and I would stop the search there. For these higher energy compounds, I am still unable to generate conformers for these, but it was a small percentage of the ~100K library that I was using.
Hi Pavlína ! I encounter same problem with you but I have no ideas what happen to it. I run a conformational search about 1200 molecules and after running about 750 molecules,the program become quite slow.Could you give me some advices？
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology
- Medical Pharmacology
- Clinical Pharmacology