Issue with refine module

[pporwal015]

Hello there!
I was trying Model Angelo's refine module with my .mrc and pdb file. I am getting error:

File "/home/******/.conda/envs/model_angelo/lib/python3.10/site-packages/model_angelo-1.0.1-py3.10.egg/model_angelo/gnn/inference.py", line 60, in infer
voxel_size = args.voxel_size
└ {'num_rounds': 3, 'crop_length': 200, 'repeat_per_residue': 3, 'esm_model': 'esm1b_t33_650M_UR50S', 'aggressive_pruning': Tru...

AttributeError: 'Args' object has no attribute 'voxel_size'

I have tried with different map and model files and still facing same error.
Please suggest something for this error.

[jamaliki]

Hi,

I think this might be a bug! Thanks for bringing it to my attention, I will investigate.

Best,
Kiarash.

[sroet]

I ran into the same bug today (also with model_angelo refine and with the same error).
(Just leaving a comment to confirm it is probably not model/map specific and to keep me updated on any solution)

[hgxy15]

Just to follow up, I'm expeiencing the same error here, the config.json contains "target_voxel_size", I manually added "voxel_size", but it does not help

[jamaliki]

Hi @hgxy15,

Could you please send the config.json that is on your system here? Before you changed it.

Best,
Kiarash.

[hgxy15]

Hi @jamaliki , thanks for the swift reply! Here's the config file, I'm not sure if this is the correct file to look at (hub/checkpoints/model_angelo_v1.0/original/config.json):
{
"standardize_mrc_args":
{
"target_voxel_size": 1.5,
"crop_z": 0,
"bfactor_to_apply": 0,
"auto_mask": false
},
"ca_infer_args":
{
"model_checkpoint": "chkpt.torch",
"bfactor": 0,
"batch_size": 4,
"stride": 16,
"dont_mask_input": true,
"threshold": 0.05,
"save_real_coordinates": false,
"save_cryo_em_grid": false,
"do_nucleotides": false,
"save_backbone_trace": false,
"save_ca_grid": false,
"crop": 6
},
"gnn_infer_args":
{
"num_rounds": 3,
"crop_length": 200,
"repeat_per_residue": 3,
"esm_model": "esm1b_t33_650M_UR50S",
"aggressive_pruning": false,
"seq_attention_batch_size": 200
}
}
thanks in advance!

[jamaliki]

@hgxy15 are you able to check if the new branch refine-mod works? You would need to do the following steps:

1. Go to the git directory for ModelAngelo
2. run git fetch && git checkout -b refine-mod
3. Activate your ModelAngelo environment and run python setup.py install

[hgxy15]

@jamaliki We just tested the refine-mode branch and are still getting the same error : File "/Share/app/Miniconda3/envs/model_angelo/lib/python3.10/site-packages/model_angelo-1.0.1-py3.10.egg/model_angelo/gnn/inference.py", line 60, in infer
voxel_size = args.voxel_size
└ {'num_rounds': 3, 'crop_length': 200, 'repeat_per_residue': 3, 'esm_model': 'esm1b_t33_650M_UR50S', 'aggressive_pruning': Tru...

AttributeError: 'Args' object has no attribute 'voxel_size'

[hgxy15]

@jamaliki Is there a way to check whether we are actually using the latest refine-mode branch ?

[jamaliki]

It doesn't seem like you are running the new branch since the version says it is using ModelAngelo 1.0.1. It should say 1.0.2. Did you make sure to reinstall by running everything in the correct conda environment ?

[stav-ros]

Hey there, I was wondering if this issue is fixed (or has come back somehow), as I'm running 1.0.9 and it seems that it is still there.

[ccgauvin94]

This issue still exists for me in 1.0.12

[jamaliki]

Hi @stav-ros and @ccgauvin94 ,

Thank you for reminding me of this, I am having a look.

Best,
Kiarash.

[joerloeffler]

Hey, I just tried to use modelangelo with relion5.0 and I received the same error message for the refine task:

voxel_size = args.voxel_size

AttributeError: 'Args' object has no attribute 'voxel_size'

is there an update on how to fix this? Thank you and sorry if I missed something!!