baker-laboratory / RoseTTAFold-All-Atom

Other
649 stars 119 forks source link

HHBlits takes ages to run! #135

Open GM-33 opened 4 months ago

GM-33 commented 4 months ago

Hi

I am struggling with running RFAA on SLURM compute node. I am able to install it as per recommendations but I dont know if it is normal for hhblists to run for so many hours!

Sun Jul 14 06:04:55 CEST 2024

It may be necessary to compile with the extra flag -L/apps/cuda/11.8.0/lib64/stubs if libcuda.so is needed.

It may be necessary to compile with the extra flag -L/apps/cuda/11.8.0/lib64/stubs if libcuda.so is needed.

/zfs/s01/z02/home/gmustafa/mambaforge /zfs/s01/z02/home/gmustafa/mambaforge/envs/RFAA/lib/python3.10/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'protein_cof_sm': Defaults list is missing _self_. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information warnings.warn(msg, UserWarning) ^MPredicting: 0%| | 0/1 [00:00<?, ?sequences/s]^MPredicting: 100%|██████████| 1/1 [00:19<00:00, 19.40s/sequences]^MPredicting: 100%|██████████| 1/1 [00:19<00:00, 19.42s/sequences] Running HHblits against UniRef30 with E-value cutoff 1e-10

it has been more than 12 hours that hhblits has been running ..

the signalp folder contains entries:

242 Jul 14 06:10 prediction_results.txt -- 17 Jul 14 06:10 output.gff3 -- 17 Jul 14 06:10 region_output.gff3 -- 0 Jul 14 06:10 processed_entries.fasta -- 933 Jul 14 06:10 output.json

at the time of writing its almost 12 hours and hhblits show no updates

also processed entries.fasta file is empty in the signalp folder!

I don't know if there is some issue with signalp or hhblits that it takes ages to finish run..

I would be thankful if I get any valuable suggestion

best regards GM

wehs7661 commented 1 month ago

Hi @GM-33, did you end up resolving the issue? I seem to have similar issues so any insights would be greatly appreciated!