search

lorenzo-rovigatti / oxDNA

A new version of the code to simulate the oxDNA/oxRNA models, now equipped with Python bindings
https://dna.physics.ox.ac.uk/
GNU General Public License v3.0
42 stars 27 forks source link

Distance between bonded neighbors exceeds acceptable values #48

Closed vivipurplr closed 1 year ago

vivipurplr commented 1 year ago

defo-CUDA-1.log defo-CUDA-1_input.txt Dear oxDNA developers, We are simulating a 3D structure consisting of 14000 nt in oxDNA.org, and I find it quite difficult to overcome this problem. We use CaDNAno to design the structure, and then using oxview to relax the structure in oxserve. After relaxtion, we uploaded the files to the oxDNA.org. We use the defalut parameters, and chose "Need relax" . But we found the simulation was not run sucessfully, and the file downloaded from Log report the the error "...Distance between bonded neighbors exceeds acceptable values...". As we noted above, we have relaxed the structure in oxserve, and we have no idea what is the mistake. We are asking for your help. Thanks in advance for any help.

lorenzo-rovigatti commented 1 year ago

Hi, maybe you should contact directly the Sulc group who manages oxDNA.org? Or you can upload here also the configuration and topology files so that I can try to simulate them and see what's going wrong.

zoombya commented 1 year ago

Hi, you can either share the problematic structures here. Or try if ox-serve would run under equilibration settings. For this you have to turn the following flag off. (it is available in both the CPU and GPU tabs) image I would actually recomend an equilibration run of 1000000 steps before you perform your simulation on oxDNA.org

vivipurplr commented 1 year ago

Hi, you can either share the problematic structures here. Or try if ox-serve would run under equilibration settings. For this you have to turn the following flag off. (it is available in both the CPU and GPU tabs) image I would actually recomend an equilibration run of 1000000 steps before you perform your simulation on oxDNA.org

image Thank you for your reply, I wonder that you mean adjust the parameter of steps to 1000000 steps in the oxserve? Bur how can I know that an equilibration run of 1000000 steps have been completed? Thank you for your help!

vivipurplr commented 1 year ago

Hi, maybe you should contact directly the Sulc group who manages oxDNA.org? Or you can upload here also the configuration and topology files so that I can try to simulate them and see what's going wrong.

Thank you for your quick reply. Can I email the configuration and topology files to you and if you can help us to simulate them? We really appreciate that.

zoombya commented 1 year ago

The value I mentioned is the default in the GPU tab, the difference between the 2 tabs is also significant, as only the GPU tab performs MD. CPU uses MC to overcome clashing particles and is usually needed only in the very beginning of your relax run. The value of 1000000 steps is the default on the GPU tab. You'll now it's complete when the structure stops updating. (for an average origami sized structure it would take at least 15 minutes) I have to implement better error handling in the console. A good indication for now that your run is failing - is you get an [Object] in the web console immidiatly after you start your simulation. I'm part of the Sulc group, if you need help with the structure you can also send it to me m.matthies@asu.edu

vivipurplr commented 1 year ago

vivipurplr

Thank you for your quick reply. We have email corresponding files to you. If there are any questions, please contract me!