- IMP-users - salilab.org

Large spheres in output rmf3 files
by Gorbea, Jose 30 Nov '20

30 Nov '20

Hi all, Upon displaying an output rmf3 file on Chimera, I am getting a number of large, radially distributed spheres as part of the model. Within the rmf3 file hierarchy, they are not part of the System or States within (meaning I can exclude these spheres from visualization by showing only "System" in the rmf3 file hierarchy panel). I wonder exactly what they mean. If anyone has experienced these before I'd appreciate any help interpreting and/or addressing this issue. Best, Jose Gorbea

2 1

Making restraints
by SoltaniKazemi, Elham 18 Nov '20

18 Nov '20

Hello, When I had the exact distance between two residues, I used something like the following to add restraints: dist_close_restraints.append(forms.gaussian(group=physical.xy_distance, feature=features.distance(modeller_atoms[res1_pos], modeller_atoms[res2_pos]), mean=re_value, stdev=distdev)) re_value is the exact distance, and I set stdev to one. Now, my problem is that I don't have the exact distance. I have something like the following: 1- The distance between (for example) residue1 and residue2 is 4.0857 with a probability of 0.66. 2- The distance between residue1 and residue2 is between 0 and 6 with a probability of 0.78 (just as an example). Does anyone know how can I add restraints when I have probability (I want to use probabilities in my restraints)?

1 0

imp_sampcon pyRMSD installation issue
by Gorbea, Jose 18 Nov '20

18 Nov '20

Hi all, I am trying to run imp_sampcon exhaust from the actin tutorial and pyRMSD is required. I tried pip installation and git cloned the repository but got the following issue on installation. """ /cm/shared/apps/python-3.7.7/lib/python3.7/distutils/dist.py:274: UserWarning: Unknown distribution option: 'install_requires' warnings.warn(msg) running install running build running build_py running build_ext building 'pyRMSD.pdbReader' extension gcc -pthread -Wno-unused-result -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -fPIC -I/cm/shared/apps/python-3.7.7/lib/python3.7/site-packages/numpy/core/include -I/cm/shared/apps/python-3.7.7/include/python3.7m -I/cm/shared/apps/python-3.7.7/include/python3.7m -c src/pdbreaderlite/PDBReader.cpp -o build/temp.linux-x86_64-3.7/src/pdbreaderlite/PDBReader.o gcc -pthread -Wno-unused-result -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -fPIC -I/cm/shared/apps/python-3.7.7/lib/python3.7/site-packages/numpy/core/include -I/cm/shared/apps/python-3.7.7/include/python3.7m -I/cm/shared/apps/python-3.7.7/include/python3.7m -c src/pdbreaderlite/PDBReaderObject.cpp -o build/temp.linux-x86_64-3.7/src/pdbreaderlite/PDBReaderObject.o In file included from /cm/shared/apps/python-3.7.7/lib/python3.7/site-packages/numpy/core/include/numpy/ndarraytypes.h:1832:0, from /cm/shared/apps/python-3.7.7/lib/python3.7/site-packages/numpy/core/include/numpy/ndarrayobject.h:12, from /cm/shared/apps/python-3.7.7/lib/python3.7/site-packages/numpy/core/include/numpy/arrayobject.h:4, from src/pdbreaderlite/PDBReaderObject.cpp:3: /cm/shared/apps/python-3.7.7/lib/python3.7/site-packages/numpy/core/include/numpy/npy_1_7_deprecated_api.h:17:2: warning: #warning "Using deprecated NumPy API, disable it with " "#define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION" -Wcpp] #warning "Using deprecated NumPy API, disable it with " \ ^~~~~~~ src/pdbreaderlite/PDBReaderObject.cpp: In function ‘PyObject* pdbreader_read(pdbreader*, PyObject*)’: src/pdbreaderlite/PDBReaderObject.cpp:87:55: warning: narrowing conversion of ‘self->pdbreader::reader->PDBReader::all_coordinates.std::vector<_Tp, _Alloc>::size<double, std::allocator<double> >()’ from ‘std::vector<double>::size_type {aka long unsigned int}’ to ‘npy_intp {aka long int}’ inside { } [-Wnarrowing] npy_intp dims[] = {self->reader->all_coordinates.size()}; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~ src/pdbreaderlite/PDBReaderObject.cpp: At global scope: src/pdbreaderlite/PDBReaderObject.cpp:142:1: error: invalid conversion from ‘const char*’ to ‘Py_ssize_t {aka long int}’ [-fpermissive] }; ^ src/pdbreaderlite/PDBReaderObject.cpp:108:5: error: invalid conversion from ‘destructor {aka void (*)(_object*)}’ to ‘printfunc {aka int (*)(_object*, _IO_FILE*, int)}’ [-fpermissive] (destructor)pdbreader_dealloc, /*tp_dealloc*/ ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ src/pdbreaderlite/PDBReaderObject.cpp:123:23: error: invalid conversion from ‘long unsigned int’ to ‘const char*’ [-fpermissive] Py_TPFLAGS_DEFAULT| Py_TPFLAGS_BASETYPE , /*tp_flags*/ src/pdbreaderlite/PDBReaderObject.cpp:142:1: error: cannot convert ‘const char*’ to ‘traverseproc {aka int (*)(_object*, int (*)(_object*, void*), void*)}’ in initialization }; ^ src/pdbreaderlite/PDBReaderObject.cpp:142:1: error: cannot convert ‘PyMethodDef*’ to ‘PyMemberDef*’ in initialization src/pdbreaderlite/PDBReaderObject.cpp:142:1: error: cannot convert ‘PyMemberDef*’ to ‘PyGetSetDef*’ in initialization src/pdbreaderlite/PDBReaderObject.cpp:139:2: error: invalid conversion from ‘initproc {aka int (*)(_object*, _object*, _object*)}’ to ‘allocfunc {aka _object* (*)(_typeobject*, long int)}’ [-fpermissive] (initproc)pdbreader_init, /* tp_init */ ^~~~~~~~~~~~~~~~~~~~~~~~ src/pdbreaderlite/PDBReaderObject.cpp:142:1: error: invalid conversion from ‘PyObject* (*)(PyTypeObject*, PyObject*, PyObject*) {aka _object* (*)(_typeobject*, _object*, _object*)}’ to ‘freefunc {aka void (*)(void*)}’ [-fpermissiv] }; ^ src/pdbreaderlite/PDBReaderObject.cpp: In function ‘PyObject* initpdbReader()’: src/pdbreaderlite/PDBReaderObject.cpp:153:9: error: return-statement with no value, in function returning ‘PyObject* {aka _object*}’ [-fpermissive] return; ^~~~~~ src/pdbreaderlite/PDBReaderObject.cpp:155:67: error: ‘Py_InitModule3’ was not declared in this scope module = Py_InitModule3("pdbReader", NULL,"Simple pdb reading"); ^ src/pdbreaderlite/PDBReaderObject.cpp:157:11: error: return-statement with no value, in function returning ‘PyObject* {aka _object*}’ [-fpermissive] return; ^~~~~~ src/pdbreaderlite/PDBReaderObject.cpp:163:1: warning: control reaches end of non-void function [-Wreturn-type] } ^ error: command 'gcc' failed with exit status 1 """ Has anyone encountered this error before? Any help would be greatly appreciated. All the best, Jose Gorbea

2 2

Defining rigid bodies
by Altair Hernández 16 Nov '20

16 Nov '20

Hi, I would like to know the appropìate way to define rigid bodies with PMI. Let's say I am working with a multimeric complex of 8 chains, grouped in 2 subcomplex of 4 chains each (subcpx1 "ABCD" and subcpx2 "EFGH"). At the same time, each subcomplex is bound in a coiled coil region, which is fixed, and I name it CorEX (coiled coils of chains "ABCD" and coiled coils of chains "EFGH"). I would like to explore different conformations of this protein complex by allowing each chain to move in the space, but imposing that coiled coil regions must be rigid (coiled coils of chain "ABCD" must move together and cannot break, e.g). My doubt is how to define this behaviour: 1- Define 3 rigid regions per chain (N-ter; CoiledCoil; C-ter) plus define each CorEX as another rigid body. 2- Define 2 rigid regions per chain (N-ter and C-ter) and then define the 2 CorEX regions. It seems that IMP.pmi does not allow to define it as in 1, some errors occur. I guess I cannot define two times a given region as a rigid body if it is already defined. I hope it is well explained :) Thanks! Altair

2 1

Suffle before sampling: fixed parts
by Altair Hernández 04 Nov '20

04 Nov '20

Hi, I am trying to sample a PDB with different restraints. I would like to shuffle only some molecules, in order to have some of them in a fixed position (the coordinates should not change after sampling). I define these rigid particles as rigid bodies before shuffling, and suffling as: IMP.pmi.tools.shuffle_configuration(root_hierarchy, excluded_rigid_bodies=fixed_particles, max_translation=50, verbose=True, cutoff=5.0, niterations=100) - Fixed particles change their coordinates after shuffling. - In the PDB (4 chains "ABCD"), the chains appear disconnected, even if I apply connectivity restraint. I may do something wrong, any help would be appreciated! Thanks! Altair

2 1

PMI with multiple PDBs
by Altair Hernández 04 Nov '20

04 Nov '20

Hi, I am trying to use multiple PDB files using the PMI multiscale modelling protocol. I'm doing it by creating different PMI.states within the main system. Is this the IMP.pmi way to do it? My script: data_directory = "../input/" pict_pdb = data_directory + "path/to/pdb1" hg_pdb_model = data_directory + "path/to/pdb2" hg_fasta = data_directory + "path/to/pdb2.fasta" # Initialize model m = IMP.Model() # HEMOGLOBIN hg_fasta_seqs = IMP.pmi.topology.Sequences(hg_fasta) hg_pdb_seqs = IMP.pmi.topology.PDBSequences(m, hg_pdb_model) # PICT PDB pict_pdb_seqs = IMP.pmi.topology.PDBSequences(m, pict_pdb) # Create System and State s = IMP.pmi.topology.System(m) # defining the root of a hierarchy st_1 = s.create_state() st_2 = s.create_state() Thanks! Best, Altair

2 1

SAXS restraint error
by Cameron Murray 21 Oct '20

21 Oct '20

Hi, I am working with the pmi library and I am trying to add an IMP.pmi.restraints.saxs.SAXSRestraint but I am getting an error. The error does not come when I add the restraint and as far as I can tell the system understands what I have given it. The issue arises when I add the restraint to the model and then try to optimize_flexible_beads(), here is the full error: Traceback (most recent call last): File "test.py", line 154, in <module> dof_obj.optimize_flexible_beads(100) File "/usr/local/lib/python3.8/site-packages/IMP/pmi/dof/__init__.py", line 518, in optimize_flexible_beads mc.optimize(nsteps) File "/usr/local/lib/python3.8/site-packages/IMP/pmi/samplers.py", line 201, in optimize self.mc.optimize(nstep * self.get_number_of_movers()) File "/usr/local/lib/python3.8/site-packages/IMP/__init__.py", line 5041, in optimize return _IMP_kernel.Optimizer_optimize(self, max_steps) _IMP_kernel.IOException: Profile is not sampled for q = 0.0131, q_max = 0.5 You can remove points with q > 0.5 from the experimental profile or recompute the profile with higher max_q I dont understand what this means, as I have set the maxq to be 0.3 in the restraint and the profile object has 0.3 as its maxq. q = 0.0131 is my first data point. I also get a similar error if I either remove the optimize_flexible_beads step or add the restraint to the model after optimization and then run replica exchange, here is that error: Traceback (most recent call last): File "test.py", line 172, in <module> rex.execute_macro() File "/usr/local/lib/python3.8/site-packages/IMP/pmi/macros.py", line 395, in execute_macro output.init_stat2(low_temp_stat_file, File "/usr/local/lib/python3.8/site-packages/IMP/pmi/output.py", line 820, in init_stat2 d = l.get_output() File "/usr/local/lib/python3.8/site-packages/IMP/pmi/restraints/__init__.py", line 107, in get_output score = self.evaluate() File "/usr/local/lib/python3.8/site-packages/IMP/pmi/restraints/__init__.py", line 83, in evaluate return self.weight * self.rs.unprotected_evaluate(None) File "/usr/local/lib/python3.8/site-packages/IMP/__init__.py", line 3785, in unprotected_evaluate return _IMP_kernel.RestraintSet_unprotected_evaluate(self, da) _IMP_kernel.IOException: Profile is not sampled for q = 0.0131, q_max = 0.5 You can remove points with q > 0.5 from the experimental profile or recompute the profile with higher max_q It seems like the issue is occuring after I add the restraint so I dont think it is a file format issue but like I said im not sure what is going on. Anything you could do to help would be appreciated. Thanks, Cameron

2 3

Issue with Flexible Modeling Using PMI
by Jake Sumner 21 Oct '20

21 Oct '20

Hello, I have recently been trying to conduct flexible modeling using beads of varying size for a particular protein. I have separated the protein into 5 domains based on the PDB crystal structure, and I have already done rigid body modeling to fit the domains to an EM map we have. However, I seem to run into an issue when I try to do flexible modeling by creating bead models using the topology file. Essentially, each of the domains is coming directly from the PDB crystal structure, and I am creating beads from the residues listed there and in the FASTA file. The beads are created with a varying number of residues to reflect the size and overall flexibility of the domain. When I set up the topology file in this manner, and make sure to not invoke the rigid body or super rigid body options, I get a structure warning from IMP saying it is making the components flexible. However, I also get a warning that "No hierarchies were passed to create_flexible_beads()". I am not sure why no flexible beads are being created from the PDB/FASTA specified residues. I am using the "ToplogyReader" and the "execute_macro" functions to set up the hierarchy and the DOF for the MC simulation. I have tried 'enable_all_movers()' to make the beads flexible, but they don't seem to exist. When I try 'dof.get_flexible_beads()' or 'dof.get_movers()', both of the lists are empty. Because the movers list is empty, I am unable to run the MC simulation macro. Could you possibly help me diagnose this problem? Is there a simple way to do the flexible bead modeling and fit it to an EM map, as I have described above? Is there some consideration I am missing in my script? Thank you, and if you need more information on the errors or the system I am working with, I am happy to provide it. Best, Jacob

2 1

Coarse Grain PDB structure
by Altair Hernández 20 Oct '20

20 Oct '20

Hi IMP users, I would like to know how to coarse a PDB structure using IMP, in a way to represent it as spheres but reducing the level of resolution (each 10 aa grouped and represented as a sphere). I have tried reading the input PDB with IMP.atom.read_pdb to generate the atom hierarchy and with IMP.atom.setup_coarse_secondary_structure_residues, however without success. Should I try using IMP.pmi instead? Thanks! Altair

2 1

Precision for systems with fixed bodies
by Atefeh Rafiei 13 Oct '20

13 Oct '20

Hi there, I have a question regarding the interpretation of the last step of integrative modeling. I am wondering if the precision calculations (for structural clusters) presented in the analysis workflow (from PMI analysis repository on GitHub), will calculate precision for all components of the modeling system including the fixed rigid bodies? I am asking this as I have a good portion of my protein system being fixed in the modeling (for specific reasons), so I am not sure if that will impact the precision I get. Thanks, Atefeh

2 1