Roden Luo

Same issue for `--model_preset=multimer`, `--use_gpu_relax=True` with `v2.2.0`.

This is also the case when `--use_gpu_relax=False`. Both stop at ``` simtk.openmm.OpenMMException: Particle coordinate is nan ```

I just got approval that I can share the following sequences for debugging purposes. In my case, this `nan` related bug happened for: ``` $ cat MPK4_MKK2_Docking.fasta >MPK4 MSAESCFGSSGDQSSSKGVATHGGSYVQYNVYGNLFEVSRKYVPPLRPIGRGAYGIVCAATNSETGEEVAIKKIGNAFDNIIDAKRTLREIKLLKHMDHENVIAVKDIIKPPQRENFNDVYIVYELMDTDLHQIIRSNQPLTDDHCRFFLYQLLRGLKYVHSANVLHRDLKPSNLLLNANCDLKLGDFGLARTKSETDFMTEYVVTRWYRAPELLLNCSEYTAAIDIWSVGCILGETMTREPLFPGKDYVHQLRLITELIGSPDDSSLGFLRSDNARRYVRQLPQYPRQNFAARFPNMSAGAVDLLEKMLVFDPSRRITVDEALCHPYLAPLHDINEEPVCVRPFNFDFEQPTLTEENIKELIYRETVKFNPQDSV >MKK2-Docking...

While waiting for the developers to respond with a new version, you can always use PyMol/ChimeraX to convert a PDB to mmCIF. Besides, you may find the below links useful....

The minimal example given by @cqc-alec is fantastic: https://github.com/cqc-alec/pybind11-3081. I turned it into a cmake one. I changed the module's name to `A_core` for clarity. The contents of the new...

Interestingly enough, it seems in my specific case there in [oxDNA repo issue #31](https://github.com/lorenzo-rovigatti/oxDNA/issues/31), I can safely remove the linking (see below). Not sure why it's not causing any linking...

Here are all files for 1bna for later references for this issue. [1bna.zip](https://github.com/BradyAJohnston/MolecularNodes/files/8644963/1bna.zip) | Name | Source | Can be loaded? | | ------------- | ------------- | ------------- | |...

It did not happen often later. So, I did not fix it.

Hi Michael, Thanks! I tried just now on both Win10 and Mac chrome. The problem still exists. I did not do anything after drag and drop the .oxview file I...

If I load [oxdna-viewer/tree/master/examples/icosahedron](https://github.com/sulcgroup/oxdna-viewer/tree/master/examples/icosahedron), and then do the following, it is indeed consistent. I didn't choose this way in the first place because this method has given me none-complete base...