jhmlam / nucleicnet Goto Github PK

Hello, I'm trying to follow this discussion, but it seems like that the output files only conclude the predicted binding score for grids but not conclude the predicted score for every residue in the protein which is necessary for the comparison between the NucleicNet and other methods. So how can I find these results?
Any help would be really appreciated. Thank you in advance!

Hi, Jordy

This is Siyuan from Prof Gao's group. I'm trying to reproduce the older version of NucleicNet, but I cannot find the old script.

We've already built a database at https://arxiv.org/abs/2205.07673

I've already finished the Feature Generation and Deep Learning procedure, I think I've encountered the error in the NucleicNet_VisualisePymol.py, thus I cannot generate the .pse file.

-rw-rw-r-- 1 chens0a g-chens0a 1.5M Oct 31 14:21 6niy_all_grids.xyz
-rw-rw-r-- 1 chens0a g-chens0a 50K Oct 31 14:21 6niy_DBScanBootstrapGrid.dict
-rw-rw-r-- 1 chens0a g-chens0a 2.0K Oct 31 14:21 6niy_dbscan_centroid_Bootstrap.xyz
-rw-rw-r-- 1 chens0a g-chens0a 59K Oct 31 14:26 6niy.pdb
-rw-rw-r-- 1 chens0a g-chens0a 116K Oct 31 14:23 6niy_strong_Bootstrap.pdb
-rw-rw-r-- 1 chens0a g-chens0a 71K Oct 31 14:21 6niy_strong_Bootstrap.xyz
-rw-rw-r-- 1 chens0a g-chens0a 1.7K Oct 31 14:26 6niy_temp.pml

PyMOL>save 5axw_pymol.pse
Save: Please wait -- writing session file...
ExecutiveGetSession: a Python error occured during creation of the session object:
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xc9 in position 2: invalid continuation byte

Do you know how to solve this problem? Hope that you can help. Thank you!

You can always find me at [email protected]

Dear professor, I am a graduate student at Central South University in China. I am recently trying to install your NucleicNet package locally. But I encounter a problem when I run it following the /NucleicNet/Notebooks07_DownstreamApplication.ipynb. The following is the python error that I got.

Traceback (most recent call last):
File "run.py", line 112, in
ServerC.MakeSXPR(aux_id=0, Checkpointlist = [ # NOTE These three models (with gelu)
File "/home/li_server3/jiapengzhen/NucleicNet/NucleicNet/Notebooks/../NucleicNet/DatasetBuilding/commandServer.py", line 735, in MakeSXPR
result_df, result_df_modelaveraged, ClassName_ClassIndex_Dict = PredictionWrapperC.Run(aux_id = aux_id)
File "/home/li_server3/jiapengzhen/NucleicNet/NucleicNet/Notebooks/../NucleicNet/DatasetBuilding/commandServer.py", line 511, in Run
benchmark_df = PredictionHelperC.Testing(ckp_hyperparam)
File "/home/li_server3/jiapengzhen/NucleicNet/NucleicNet/Notebooks/../NucleicNet/DatasetBuilding/commandServer.py", line 329, in Testing
test_benchmark = self.Diagnose(Testing_PdbidBatches, ckp_hyperparam)
File "/home/li_server3/jiapengzhen/NucleicNet/NucleicNet/Notebooks/../NucleicNet/DatasetBuilding/commandServer.py", line 169, in Diagnose
ds_ = self.FetchDatasetC.GetNoOpSingleDataset(
AttributeError: 'FetchDataset' object has no attribute 'GetNoOpSingleDataset'

I find that the class FetchDataset at /NucleicNet/DatasetBuilding/commandDataFetcher.py has no attribute 'GetNoOpSingleDataset'.
Could you help me to solve it?
Thanks very much.

chmod +x should be applied recursively on those folders

Dear professor, I am a graduate student at Shandong University in China. I am recently doing research on protein-RNA binding sites and read your paper "A deep learning framework to predict binding preference of RNA constituents on protein surface". This is a very great work and I am very inspired by your team's research and the design of "NucleicNet".
I want to get the datasets and labels of your research to continue my work, could you please send me ([email protected])? Thank you very much!