DeepLearningExamples/DGLPyTorch/DrugDiscovery/RoseTTAFold/folding/README
2021-10-15 15:46:41 +02:00

44 lines
1.7 KiB
Plaintext

###############################################################################
RosettaTR.py: Relax on dualspace (2 rounds of relax w/ different set of
restraints)
###############################################################################
usage: RosettaTR.py [-h] [-r NRESTARTS] [-pd PCUT] [-m {0,1,2}] [-bb BB]
[-sg SG] [-n STEPS] [--save_chk] [--orient] [--no-orient]
[--fastrelax] [--no-fastrelax] [--roll] [--no-roll]
NPZ FASTA OUT
positional arguments:
NPZ input distograms and anglegrams (NN predictions)
FASTA input sequence
OUT output model (in PDB format)
optional arguments:
-h, --help show this help message and exit
-r NRESTARTS number of noisy restrarts (default: 3)
-pd PCUT min probability of distance restraints (default: 0.05)
-m {0,1,2} 0: sh+m+l, 1: (sh+m)+l, 2: (sh+m+l) (default: 2)
-bb BB predicted backbone torsions (default: )
-sg SG window size and order for a Savitzky-Golay filter (comma-
separated) (default: )
-n STEPS number of minimization steps (default: 1000)
--save_chk save checkpoint files to restart (default: False)
--orient use orientations (default: True)
--no-orient
--fastrelax perform FastRelax (default: True)
--no-fastrelax
--roll circularly shift 6d coordinate arrays by 1 (default: False)
--no-roll
# USAGE
conda activate folding
# try: -m 0,1,2
# -pd 0.05, 0.15, 0.25, 0.35, 0.45
# repeat ~3-5 times for every combination of -m and -pd
# !!! use '--roll' option if no-contact bin is the last one !!!
python ./RosettaTR.py -m 0 -pd 0.15 fake.npz fake.fa model.pdb