Roden Deng Luo

Hi, I frequently face errors without much info. The call stack has only main and the algorithm runner (attached). This happens when there is something wrong in the Python plugin I developed. I wonder what is your development environment. Can I debug C++ and Python at the same time? It is fine if the whole process involves multiple IDEs or commandlines. Right now, if there is anything wrong in Python, I can only insert many print() to figure out why. Any help is much appreciated. Best, Roden

Hi, I wonder how open and what is the community development practice for Tulip (given most threads I see are quite outdated). And this is my first time thinking of developing on top of a SourceForge project, which I must say a great tool. I get a few questions. If I develop something can I merge to the main code base and will there be a reviewer? Can I get any (resonable amount) support? Are you thinking of moving to GitHub for example? Specifically, at the moment, I am adding more python bindings...

Tulip 6.0.0 build errors and solutions on WSL2 Ubuntu 24.04.2 LTS

Hi Lorenzo, Thanks very much! I will decrease dt in the next round if I meet it again. I have linked to this issue from the protein integrated simulator anm-oxdna's github repo.

Hi, I run into this many times. And it is after running "a few steps" I get the error. Following the above suggestion, usually after I increase the "max_density_multiplier" or decrease the box size, or decrease the number of strands/nucleotides, it will work. If it does not work, then it is probably hitting the upper limit of my memory (Core Dump/Segmentation fault), so I give up with that simulation setting. One thing to note is that, if I do not use external force, I seldom encountered this error....

After I set back_in_box=true, I still see strands out of the bouding box that oxdna-viewer draws (PBC=None). But they do stay close to the cell in the center after a while of simulation. (It does not look like the one I shared two posts before, "back_in_box-false_PBC-None.png"). It looks to me there is a larger bounding box. The reason I am trying different parameter combinations is, I want to find one that can keep the "double-strand helix" with the visual glitch (the ripped backbones) removed....

Sure. It is a 17G trajectory file. In case you want to reproduce a similar error, you will probably need a fairly large trajectory file and set the max_io small. (I still got some questions related to back_in_box, I am draftting them in the main thread.) INFO: Setting the random number generator with seed = 1056332879 INFO: The generator will try to take into account bonded interactions by choosing distances between bonded neighbours no larger than 2.000000 INFO: Running Debye-Huckel at salt concentration...

Hi Lorenzo, Thanks! I tried the script just now. It run through after I set max_io=100.0. If anyone experience where the converted trajectory has less time steps than the original one, there might be an error even though the program ended like normal. The error log is in log.dat. (Probably because I have not been that familar with oxDNA flow yet. It took me a while to locate the problem. (base) [luod@gpu210-10 back_in_box]$ DNAnalysis input_back_in_box (base) [luod@gpu210-10 back_in_box]$ echo $?...