rosettacommons / pyrosetta.notebooks Goto Github PK

Is there a new code for initializing PyRosetta on Google Colab?
I keep receiving an error that it is already mounted, so then I told it to force re-mount and then I get an error about the %%time, and then I get an error about Python 3.7
Everything was working a couple of months ago, so just wanted to check to see if there is some update I am unaware of.
Thank you!

Hi,

I am using Pyrosetta to model different peptides in low and high resolution with algorithms such as CCD and KIC and so far it is working fine. Even though the results I am getting are fine, I wanted to know more about what Pyrosetta is doing in the background, so I started looking into the info that is printed during the modelling but it seems that I can only access data like the inner/outer cycles and the score calculated after every cycle. Is it possible to access more specific data about my modelling?

Here is an example of the info I can access;

protocols.loops.loop_mover.refine.LoopMover_Refine_CCD: cycle: 1 -300.2965
protocols.loops.loop_mover.refine.LoopMover_Refine_CCD: refinement cycle (outer/inner): 1/5 1/10
protocols.loops.loop_mover.refine.LoopRefineInnerCycle: Setting up data for LoopRefineInnerCycleContainer.
protocols.loops.loop_mover.refine.LoopRefineInnerCycle: Setting up data for SmallMinCCDTrial.
core.pack.task: Packer task: initialize from command line()
core.pack.task: Packer task: initialize from command line()
protocols.moves.MonteCarlo: MonteCarlo:: last_accepted_score,lowest_score: -340.47098 -340.47098
protocols.loops.loop_mover.refine.LoopRefineInnerCycle: Setting up data for ShearMinCCDTrial.
core.pack.task: Packer task: initialize from command line()
core.pack.task: Packer task: initialize from command line()

Thank you in advance.

Kind regards,

Roberto

Hello everyone!
For the first time using Pyrosetta on a Linux server, I first created a new environment using anaconda. After entering this newly created environment, I used the conda install pyrosett command to successfully install pyrosetta, and then I successfully used pycharm's SSH interpreter locally. Loaded the python interpreter in this remote environment, but there were some problems when writing the program, for example, the reference “protocols” could not be found in the “init.py”, the unresolved reference “MinMover”, the unresolved reference 'ScoreFunction' and other such problems, of course, I also tried to start jupyter notebook in the terminal to write the program, this is no problem, but jupyter notebook is not convenient to debug the program, so I want to consult, have any friends encountered Same problem as me? Can share the solution.

Hello,

I just followed the instructions to download and get running on Colab and while everything seemed to install properly with the exception of the following error:
ERROR: dask-jobqueue 0.7.1 has requirement distributed>=2.5.2, but you'll have distributed 1.25.3 which is incompatible.

I get the following error when executing notebook 1.02:

ModuleNotFoundError: No module named 'pyrosetta.distributed.utility.pickle'

Any suggestions? Thanks.

Scott

Can I ask these questions here?

I am currently working on the PyRosetta workshops and had a few questions regarding the notebook titled "Membrane Protein ∆∆G of Mutation" (notebooks/15.02-Membrane-Protein-ddG-of-mutation.ipynb) available in the PyRosetta GitHub repository: https://github.com/RosettaCommons/PyRosetta.notebooks/blob/master/notebooks/15.02-Membrane-Protein-ddG-of-mutation.ipynb

I encountered an error when using the 3gp6.pdb file that I downloaded. As an alternative, I downloaded the 1mm4.pdb file mentioned in the research paper. However, the REU values obtained from 1mm4.pdb were unusually high. Eventually, I found the 3gp6_A.pdb file located at notebooks/inputs/3gp6_A.pdb, which yielded faster and consistent results. Could you please explain how the 3gp6_A.pdb file was prepared to better correlate with the data?

Regarding the cutoff of 1.5 REU, I obtained the same graph but I could not Identify the amino acids. Did you calculate it as PREDICTED_VALUES - EXPERIMENTAL_VALUES?

Did you convert the REU to kcal/mol?
Thank you for your assistance!

Hi team,

Thanks for a great resource. I've been struggling for a few days to troubleshoot this issue. I'm getting a very cryptic index error when trying to run the redesign Chain A in this notebook.

I have tried running with 2022 (python 3.9) and 2019 (python 3.6) releases of PyRosetta installed via conda, but the error persists.

I'd be grateful for any suggestions.

All the best,

Konrad.

Denovo Protein Design Notebook Output.pdf

Notebook 15.1 - accounting for the lipid membrane is failing in master with a timeout. We can put a debug statement around creating the bilayer, but perhaps there is a way to make it faster instead of knocking out the tested components?

Hi PyRosetta team,

I encountered a Python version conflict issue when initializing PyRosetta in google colab (when running: from pyrosetta import *
and init() code block). I was wondering if there is a way to solve it. Please find the below details for this conflict. Thank you so much.

ImportError: Python version mismatch: module was compiled for Python 3.6, but the interpreter version is incompatible: 3.7.10 (default, Feb 20 2021, 21:17:23)
[GCC 7.5.0].

when I set the initial parameter as
init("-use_input_sc -ex1 -ex2 -nstruct 1 -out:path:pdb ./tutorial_output -out:path:score ./expected_output")
it output nothing, how can I get the pdb file result?

I want to know whether the following ERROR/warning matters when I mutate amino acids in pyrosetta:

core.import_pose.import_pose: File '/data2/rjli/test_dataset/1amk/1amk_protein.pdb' automatically determined to be of type PDB
core.io.pdb.HeaderInformation: [ WARNING ] Deposition day not in range [1, 31]: 0
core.io.pdb.HeaderInformation: [ WARNING ] Unrecognized month in HEADER deposition date             
core.io.pdb.HeaderInformation: [ ERROR ] Malformed Compound record found: ' 1AMK_PROTEIN'
core.io.pose_from_sfr.PoseFromSFRBuilder: [ WARNING ] discarding 4 atoms at position 1 in file /data2/rjli/test_dataset/1amk/1amk_protein.pdb. Best match rsd_type:  SER:NtermProteinFull

THE PYTHON CODE :

import os
from pyrosetta import *
from pyrosetta.rosetta.protocols.simple_moves import MutateResidue
init()

source_folder = '/data2/rjli/test_dataset'

for id in os.listdir(source_folder):
    folder_path = os.path.join(source_folder, id)
    if os.path.isdir(folder_path):
        for file in os.listdir(folder_path):
            if file.endswith('protein.pdb'):
                protein_file = os.path.join(folder_path, file)
                pose = pose_from_pdb(protein_file)
                n_res = pose.total_residue()
                # loop over all residues
                for i in range(1, n_res + 1):
                    res = pose.residue(i)
                    if res.name() == "ASN":
                        # create a MutateResidue mover to mutate Asn to Leu
                        mut = MutateResidue(i, "LEU")
                        mut.apply(pose)
                    if res.name() == "GLN":
                        # create a MutateResidue mover to mutate GLN to Leu
                        mut = MutateResidue(i, "LEU")
                        mut.apply(pose)
                output_file = os.path.join(folder_path, f'{id}_mutate0_protein.pdb')
                pose.dump_pdb(output_file)

The function signature has changed for the parser and this needs to be fixed. Also, debug statements should only not run for long components. Not sure where that part changed. I have a branch with this fixed, but for some reason, can't push to the repo.

when I use pyrosetta to convert protein amino acid, there is some amino acid keep their original state, some others converted as I commanded.

The function I used is MutateResidue(). As the picture below, the LEU is a converted AA from ASN, but there is still one ASN cannot to be concerted into LEU.

antibody_designer.linuxgccrelease -s my_ab.pdb -seq_design_cdrs L1 L2 H3 \ -graft_design_cdrs L1 L2 -light_chain kappa -design_H3_stem -inner_kt 2.0 \ -outer_kt 2.0 -seq_design_profile_samples 5 -nstruct 5 -out:prefix tutD5_ <AntibodyDesignMover name="RAbD" seq_design_cdrs="L1,L3,H1,H3" mintype="relax" inner_kt="2.0" outer_kt="2.0"/>
https://nbviewer.org/github/RosettaCommons/PyRosetta.notebooks/blob/master/notebooks/12.02-RosettaAntibodyDesign-RAbD.ipynb

This example should be executed in rosetta, what I want to know is how to execute this command in pyrosetta. Thanks

Below is the test.py script for my current environment

@ajasja @momeara @jueseph @pjsteiner @danielvarela

Dear Pyrosetta developers,

I am recently trying to do an analysis on protein dihedral angles.

I have all the dihedral angle atom information (their atom type, atom name, atom coordinates, and index from PDB) saved in a dictionary and I am able to calculate each dihedral angle (not only backbone ones, also improper ones) using packages like MDAnalysis.

Then I am adding a small change to each of the dihedral angles to see how this will influence the conformation of the original protein structure, which requires me to convert the modified dihedral angles back to cartesian coordinates, is there any way in Pyrosetta I can do this?

Thanks for any suggestions!

Best,
Tianming

Hi team,

May I ask how to add the acetylated N-terminus and methylamidated C-terminus to a pose when using Pyrosetta? The pose is loaded from a pdb using a command like:
pose = pose_from_pdb("test.pdb").

I googled as much as I possible, but still cann't find a command that can work.
There was a command seemed to worked, but still it didn't:
pyrosetta.rosetta.core.conformation.add_variant_type_to_conformation_residue(pose.conformation(), pyrosetta.rosetta.core.chemical.ACETYLATED_NTERMINUS_VARIANT, 1)

The error was:
ERROR: Unable to find desired residue 'LYS:NtermProteinFull' with variant 'ACETYLATED_NTERMINUS_VARIANT'. Attempted to add target variant(s) to ResidueType using both ResidueType base name 'LYS' and base ResidueType. Was attempting to add new variant type 'ACETYLATED_NTERMINUS_VARIANT'

Thank you very much.

All the best.

import os
print (os.environ['CONDA_DEFAULT_ENV'])

prot_design
!mamba list
packages in environment at /home/ubuntu/miniconda3/envs/prot_design:

Name Version Build Channel
_libgcc_mutex 0.1 conda_forge conda-forge
_openmp_mutex 4.5 2_gnu conda-forge
bzip2 1.0.8 h7f98852_4 conda-forge
ca-certificates 2022.12.7 ha878542_0 conda-forge
environment-kernels 1.2.0 pypi_0 pypi
ld_impl_linux-64 2.39 hcc3a1bd_1 conda-forge
libffi 3.4.2 h7f98852_5 conda-forge
libgcc-ng 12.2.0 h65d4601_19 conda-forge
libgomp 12.2.0 h65d4601_19 conda-forge
libnsl 2.0.0 h7f98852_0 conda-forge
libsqlite 3.40.0 h753d276_0 conda-forge
libuuid 2.32.1 h7f98852_1000 conda-forge
libzlib 1.2.13 h166bdaf_4 conda-forge
ncurses 6.3 h27087fc_1 conda-forge
openssl 3.0.7 h0b41bf4_1 conda-forge
pip 22.3.1 pyhd8ed1ab_0 conda-forge
pyrosetta 2022.49+release.201d7639f91 pypi_0 pypi
python 3.9.15 hba424b6_0_cpython conda-forge
readline 8.1.2 h0f457ee_0 conda-forge
setuptools 65.5.1 pyhd8ed1ab_0 conda-forge
tk 8.6.12 h27826a3_0 conda-forge
tzdata 2022g h191b570_0 conda-forge
wheel 0.38.4 pyhd8ed1ab_0 conda-forge
xz 5.2.6 h166bdaf_0 conda-forge

!which conda
/home/ubuntu/miniconda3/condabin/conda

!which python
/home/ubuntu/miniconda3/envs/prot_design/bin/python

# python -c "import pyrosetta; pyrosetta.init()"
# import pyrosetta
from pyrosetta.rosetta import *

ModuleNotFoundError Traceback (most recent call last)
Cell In [6], line 3
1 # python -c "import pyrosetta; pyrosetta.init()"
2 # import pyrosetta
----> 3 from pyrosetta.rosetta import *

ModuleNotFoundError: No module named 'pyrosetta'

cdr_selector = CDRResidueSelector()

ERROR: Assertion active( key ) failed.
ERROR:: Exit from: /home/benchmark/rosetta/source/build/PyRosetta/linux/clang-3.4.2/python-3.8/release/source/src/utility/keys/SmallKeyVector.hh line: 548
Traceback (most recent call last):
File "", line 1, in
RuntimeError:

File: /home/benchmark/rosetta/source/build/PyRosetta/linux/clang-3.4.2/python-3.8/release/source/src/utility/keys/SmallKeyVector.hh:548
[ ERROR ] UtilityExitException
ERROR: Assertion active( key ) failed.

the code using dask with pyrosetta did not work. in the last two step. it raise AttributeError: module 'pyrosetta.rosetta' has no attribute 'path'. please check it.

For example, I want to use python to customize a dynamic custom scoring function on top of the scoring function that comes with pyrosetta, and I've looked at the tutorials and can't seem to find a clear way to solve this problem. A clear answer would be greatly appreciated.

What is really missing is an example of snugdock usage with pyrosetta

I am using pyrosetta to mutate specific amino acids to output a new protein file, but I encounter the following trouble when it comes to (pdbid 3V4T) which contains HETATM IAS :

core.conformation.Conformation: [ WARNING ] missing heavyatom:  OXT on residue GLU:CtermProteinFull 418
core.conformation.Conformation: [ WARNING ] missing heavyatom:  OXT on residue GLU:CtermProteinFull 836
core.conformation.Conformation: [ WARNING ] missing heavyatom:  OXT on residue GLU:CtermProteinFull 1254
core.conformation.Conformation: [ WARNING ] missing heavyatom:  OXT on residue GLU:CtermProteinFull 1672
core.conformation.Conformation: [ WARNING ] missing heavyatom: N on residue pdb_IAS 1673
core.conformation.Conformation: [ WARNING ] missing heavyatom: CA on residue pdb_IAS 1673
core.conformation.Conformation: [ WARNING ] missing heavyatom: C on residue pdb_IAS 1673
core.conformation.Conformation: [ WARNING ] missing heavyatom: O on residue pdb_IAS 1673
core.conformation.Conformation: [ WARNING ] missing heavyatom: CB on residue pdb_IAS 1673
core.conformation.Conformation: [ WARNING ] missing heavyatom: CG on residue pdb_IAS 1673
core.conformation.Conformation: [ WARNING ] missing heavyatom: OD1 on residue pdb_IAS 1673
core.conformation.Conformation: [ WARNING ] missing heavyatom: OD2 on residue pdb_IAS 1673
core.conformation.Conformation: [ WARNING ] Rebuilding missing atom (HXT) from residue pdb_IAS 1673 from faked coordinates, due to too few atoms being present.
This probably means that this residue should be further optimized...

ERROR: too many tries in fill_missing_atoms!
ERROR:: Exit from: /home/benchmark/rosetta/source/src/core/conformation/Conformation.cc line: 1933
Traceback (most recent call last):
  File "/data2/rjli/ROSETTA_SCRIPTS/Disable_H-bond.py", line 15, in <module>
    pose = pose_from_pdb(protein_file)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/data2/rjli/mambaforge/envs/rosetta/lib/python3.11/site-packages/pyrosetta/io/__init__.py", line 20, in pose_from_pdb
    return pose_from_file(filename)
           ^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: 

File: /home/benchmark/rosetta/source/src/core/conformation/Conformation.cc:1933
[ ERROR ] UtilityExitException
ERROR: too many tries in fill_missing_atoms!

Does anyone have advice for how to resolve this issue? Am I missing something basic. I am using the academic license provided from the University of Washington that was emailed.

Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/
Requirement already satisfied: pyrosettacolabsetup in /usr/local/lib/python3.8/dist-packages (1.0.6)
Drive already mounted at /content/google_drive; to attempt to forcibly remount, call drive.mount("/content/google_drive", force_remount=True).
Looking for compatible PyRosetta wheel file at google-drive/PyRosetta/colab.bin/wheels...
To obtain PyRosetta license please visit https://www.rosettacommons.org/software/license-and-download
Please enter you RC license login:
Please enter you RC license password:
Downloading PyRosetta package...
--2023-01-22 04:08:09-- https://west.rosettacommons.org/pyrosetta/release/release/PyRosetta4.MinSizeRel.python38.ubuntu.wheel/.latest

HTTP request sent, awaiting response... 401 Unauthorized
Authentication selected: Basic realm="Rosetta Repo"

HTTP request sent, awaiting response... 401 Unauthorized

ERROR: Wheel download has not worked correctly, ABORTING...

An exception has occurred, use %tb to see the full traceback.

The following two sections are missing.

Chapter 1.0 How to Get Started
1.1 PyRosetta Google Drive Setup
1.2 PyRosetta Google Drive Usage Example

But maybe there's a reason for that? :/

System info:

• transformers version: 4.37.1
• Platform: Linux-4.18.0-513.11.1.el8_9.x86_64-x86_64-with-glibc2.28
• Python version: 3.9.16
• Huggingface_hub version: 0.20.2
• Safetensors version: 0.3.3
• Accelerate version: 0.26.1
• Accelerate config: not found
• PyTorch version (GPU?): 2.1.2 (True)
• Tensorflow version (GPU?): not installed (NA)
• Flax version (CPU?/GPU?/TPU?): 0.7.5 (cpu)
• Jax version: 0.4.23
• JaxLib version: 0.4.23
• Using GPU in script?: Yes
• Using distributed or parallel set-up in script?: No

Code:

import pyrosetta
from transformers import AutoTokenizer, EsmForProteinFolding

Expected behaviour: the module imports without issue.
I can import pyrosetta on its own without issue. I can import the transformers modules without issue and run inference on PDB modules, as described in the protein structure prediction Jupyter notebook. I can do this without issue in a separate script. It is only when I import both that the segmentation fault occurs. The import order does not matter. Given that both work separately, I would expect them to work together as I cannot find any package conflicts.

Dear Madam/Sir,

I am trying to understand the docking classes outlined here.

How can I find out which classes use high resolution docking (full atomistic) and which use low resolution (centroid)? I am particularly interested in DockMCMProtocol and DockingEnsemble.

Furthermore I am interested in performing Flexible Docking using the ensemble1 and ensemble2 flags, is DockingProtocol able to do this? If so does it take care of the prepack step as well?

I Calculating psi angle using vector, angle is 111.200012, not same with 109 using pose.conformation().bond_angle(N28,CA28,C28). my cal_angle function is:
def calculate_angle(vector_ba, vector_bc):
...: dot_product = sum(a * b for a, b in zip(vector_ba, vector_bc))
...: magnitude_ba = math.sqrt(sum(a2 for a in vector_ba))
...: magnitude_bc = math.sqrt(sum(b2 for b in vector_bc))
...: angle_radians = math.acos(dot_product / (magnitude_ba * magnitude_bc))
...: angle_degrees = math.degrees(angle_radians)
...: return angle_degrees

Can you give me the cause why difference make? Thanks

I saw david baker's work on building a large ring from scratch in cell, which made me very interested. I looked at the code provided in their attachment materials. It is obvious that it is not compatible with pyrosetta 4 now. Multiple functions and classes are no longer the same. This caused me great pain when trying to reproduce it. I humbly request you to write a tutorial on how to build a cyclic peptide.

In the third antibody design notebook, the AntibodyDesignMover rabd2 crashes when applied, giving the error message:

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<ipython-input-7-2dc5bb52b04a> in <module>
     10 rabd2.set_light_chain("kappa")
     11 if not os.getenv("DEBUG"):
---> 12     rabd2.apply(pose)

RuntimeError: 

File: /home/benchmark/rosetta/source/src/protocols/antibody/database/AntibodyDatabaseManager.cc:367
[ ERROR ] UtilityExitException
ERROR: Antibody Design requires the identity of the light chain gene. Please set it using the -light_chain option.  PyIgClassify can be used to determine the light chain gene.

XmlObjects calling the AntibodyDesignMover work fine, only the python AntibodyDesignMover seems to crash. Looking up PyIgClassify in pyrosetta API reference doesn't give any result. Setting -light_chain kappa during pyrosetta.init fixes this, so I fear the problem comes from PyRosetta.

Using PyRosetta version 2020.39+release.9ee477a. The required version is 2020.38 in environment.yml, so it would be worth checking with this one and a more recent PyRosetta release.

As the notebook said, FastRelax() took up a huge amount of resource and running it crashed the notebook.
Hence, I tried it via a python script
python relax.py
The code and input are the same as that in the notebook. (From Loading structures to relax.apply())
It crashed in the step of relax.apply(testPose) and returned "Segmentation fault"

I used gdb to check the core file and got
Program terminated with signal SIGSEGV, Segmentation fault. #0 0x00007fbf718e595a in ?? ()

Env
WSL2 Ubuntu 20.04 LTS (I also tried WSL1 before, and the problem was the same)
Using conda to install pyrosetta (Release 2020.28+release)
CPU i7 7700k RAM 32 G
I also have tried ulimit -s unlimited to set the stack size.

Thank you.

I think I got it. I forget to comment out relax = rosetta.protocols.relax.FastRelax()