IMP-users

imp-users@salilab.org

274 discussions

IMP 2.4.0 release
by Ben Webb 02 Mar '15

02 Mar '15

IMP 2.4.0 is now available. This is a bug fix and minor update release. Most notably Python 3 is now supported, and command line tools ("applications") are now part of regular IMP modules. See https://github.com/salilab/imp/releases/tag/2.4.0 for the full change log. Check it out at http://integrativemodeling.org/ Ben -- ben(a)salilab.org http://salilab.org/~ben/ "It is a capital mistake to theorize before one has data." - Sir Arthur Conan Doyle

1 0

Problems compiling IMP on mac OS X version 10.9.5
by Marco Di Stefano 30 Jan '15

30 Jan '15

Dear IMP users, I am having problems in compiling IMP on my mac (OS X version 10.9.5). These are the steps I followed in some details and I hope you can help me to fix my problem. 1 - I downloaded IMP from github with this command in a directory in my home folder ~/IMP git clone git@github.com:salilab/imp.git 2 - I renamed the downloaded imp directory as src mv imp/ src 3 - In the src folder I executed the python script ./setup_git.py 4 - In ~/IMP I created the directory build-fast and from the latter directory I launched . ../prep_env_fast.sh where prep_env_fast.sh contains the following: cmake ../src/ -G Ninja -DIMP_DISABLED_MODULES=domino:multifit:system:example:misc:em3d:modeller:membrane -DCMAKE_POLICY_DEFAULT_CMP0054=OLD -DCMAKE_POLICY_DEFAULT_CMP0042=OLD -DCM\ AKE_BUILD_TYPE=Release -DIMP_MAX_LOG=SILENT -DIMP_MAX_CHECKS=NONE -DCMAKE_INSTALL_PREFIX=$HOME -DIMP_PER_CPP_COMPILATION="core:isd" It is meant to run the cmake command excluding some of the modules and tuning some other flags. The output of the script is: -- The C compiler identification is AppleClang 6.0.0.6000056 -- The CXX compiler identification is AppleClang 6.0.0.6000056 -- Check for working C compiler using: Ninja -- Check for working C compiler using: Ninja -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Check for working CXX compiler using: Ninja -- Check for working CXX compiler using: Ninja -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Running check_common_problems -- Running clean_build_dir -- Running setup_cmake -- Enabling clang C++11 support -- Wrong doxygen version at /usr/local/bin/doxygen. Found 1.8.9.1, but needed 1.8.6 -- core is percpp -- isd is percpp -- Python binary is /usr/bin/python (version 2.7.5) -- Found PythonLibs: /usr/lib/libpython2.7.dylib (found suitable exact version "2.7.5") -- Running setup -- Running setup_all -- Running making containers -- Running setup_module_alias for kernel -- Setting up alias for module kernel as -- Boost version: 1.57.0 -- Found the following Boost libraries: -- system -- filesystem -- thread -- program_options -- Performing Test IMP_COMPILER_HAS_AUTO -- Performing Test IMP_COMPILER_HAS_AUTO - Success -- Performing Test IMP_COMPILER_HAS_DEBUG_VECTOR -- Performing Test IMP_COMPILER_HAS_DEBUG_VECTOR - Failed -- Found GPERFTOOLS -- Performing Test GPerfTools_found -- Performing Test GPerfTools_found - Success -- Found GPerfTools -- Found LOG4CXX -- Performing Test Log4CXX_found -- Performing Test Log4CXX_found - Success -- Found Log4CXX -- Performing Test TCMalloc_HeapChecker_found -- Performing Test TCMalloc_HeapChecker_found - Failed TCMalloc_HeapChecker not found -- Performing Test TCMalloc_HeapProfiler_found -- Performing Test TCMalloc_HeapProfiler_found - Failed TCMalloc_HeapProfiler not found -- Running generate paths.cpp Configuring module base ... yes -- Running setup_swig_wrappers base -- Running IMP.base making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/base.h Configuring module benchmark ... yes -- Running setup_swig_wrappers benchmark -- Running IMP.benchmark making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/benchmark.h -- Build type: Release -- USING CXXFLAGS = ' -w -O3 -DNDEBUG' -- USING EXEFLAGS = ' -framework CoreFoundation ' -- Targetting Ninja -- Using /usr/bin/c++ compiler. -- DARWIN_VERSION=13 -- Mac Leopard detected -- Requested component: MPFR -- Requested component: GMP Configuring module cgal ... yes -- Running setup_swig_wrappers cgal -- Running IMP.cgal making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/cgal.h Configuring module kmeans ... yes -- Running setup_swig_wrappers kmeans -- Running IMP.kmeans making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/kmeans.h Configuring module module_template ... yes -- Running setup_swig_wrappers module_template -- Running IMP.module_template making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/module_template.h -- Found MPI_C: /usr/local/Cellar/open-mpi/1.8.4/lib/libmpi.dylib -- Found MPI_CXX: /usr/local/Cellar/open-mpi/1.8.4/lib/libmpi_cxx.dylib;/usr/local/Cellar/open-mpi/1.8.4/lib/libmpi.dylib -- MPI found Configuring module mpi ... yes -- Running setup_swig_wrappers mpi -- Running IMP.mpi making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/mpi.h Configuring module test ... yes -- Running setup_swig_wrappers test -- Running IMP.test making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/test.h -- Found ANN -- Performing Test ANN_found -- Performing Test ANN_found - Success -- Found ANN Configuring module algebra ... yes -- Running setup_swig_wrappers algebra -- Running IMP.algebra making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/algebra.h Configuring module kernel ... yes -- Running setup_swig_wrappers kernel -- Running IMP.kernel making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/kernel.h Configuring module display ... yes -- Running setup_swig_wrappers display -- Running IMP.display making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/display.h -- Found GSL -- Performing Test GSL_found -- Performing Test GSL_found - Success -- Found GSL Configuring module gsl ... yes -- Running setup_swig_wrappers gsl -- Running IMP.gsl making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/gsl.h Configuring module parallel ... yes -- Running setup_swig_wrappers parallel -- Running IMP.parallel making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/parallel.h -- Found HDF5 -- Performing Test HDF5_found -- Performing Test HDF5_found - Success -- Found HDF5 Configuring module score_functor ... yes -- Running setup_swig_wrappers score_functor -- Running IMP.score_functor making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/score_functor.h Configuring module statistics ... yes -- Running setup_swig_wrappers statistics -- Running IMP.statistics making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/statistics.h Configuring module core ... yes -- Running setup_swig_wrappers core -- Running IMP.core making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/core.h Configuring module container ... yes -- Running setup_swig_wrappers container -- Running IMP.container making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/container.h modeller is disabled Different data/build_info/sorted_modules -modeller Module IMP.modeller disabled Configuring module symmetry ... yes -- Running setup_swig_wrappers symmetry -- Running IMP.symmetry making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/symmetry.h Configuring module atom ... yes -- Running setup_swig_wrappers atom -- Running IMP.atom making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/atom.h Configuring module pmi ... yes -- Running setup_swig_wrappers pmi -- Running IMP.pmi making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/pmi.h Configuring module scratch ... yes -- Running setup_swig_wrappers scratch -- Running IMP.scratch making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/scratch.h Configuring module em ... yes -- Running setup_swig_wrappers em -- Running IMP.em making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/em.h Configuring module kinematics ... yes -- Running setup_swig_wrappers kinematics -- Running IMP.kinematics making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/kinematics.h misc is disabled Different data/build_info/sorted_modules -misc Module IMP.misc disabled -- Building internal RMF -- Boost version: 1.57.0 -- Found the following Boost libraries: -- system -- filesystem -- thread -- program_options -- iostreams -- Found HDF5: /usr/local/lib/libhdf5.dylib;/usr/local/lib/libhdf5.dylib;/usr/local/Cellar/hdf5/1.8.14/lib/libhdf5.dylib;/usr/local/opt/szip/lib/libsz.dylib;/usr/lib/libz.dylib;/usr/lib/libdl.dylib;/usr/lib/libm.dylib -- Found HDF5: /usr/local/lib/libhdf5.dylib;/usr/local/lib/libhdf5.dylib;/usr/local/Cellar/hdf5/1.8.14/lib/libhdf5.dylib;/usr/local/opt/szip/lib/libsz.dylib;/usr/lib/libz.dylib;/usr/lib/libdl.dylib;/usr/lib/libm.dylib (Required is at least version "1.8") -- Deprecated backends (and hdf5) are enabled -- Found Log4CXX -- Enabling clang C++11 support -- Running Making RMF/HDF5.h -- Updating /Users/marco/IMP/build-fast/src/dependency/RMF/include/RMF/HDF5.h -- Running Making RMF.h -- Updating /Users/marco/IMP/build-fast/src/dependency/RMF/include/RMF.h -- Running setup_cmake.py -- Running Running setup_cmake.py -- Logging enabled -- Using VMD headers found in /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/plugins/vmd/include -- Found SWIG: swig (found suitable version "3.0.2", minimum required is "1.3.40") -- Wrong doxygen version at /usr/local/bin/doxygen. Found 1.8.9.1, but needed 1.8.6 -- RMF doc disabled Configuring module rmf ... yes -- Running setup_swig_wrappers rmf -- Running IMP.rmf making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/rmf.h Configuring module rotamer ... yes -- Running setup_swig_wrappers rotamer -- Running IMP.rotamer making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/rotamer.h Configuring module saxs ... yes -- Running setup_swig_wrappers saxs -- Running IMP.saxs making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/saxs.h -- Found HDF5 -- Found HDF5 domino is disabled Different data/build_info/sorted_modules -domino Module IMP.domino disabled Configuring module isd ... yes -- Running setup_swig_wrappers isd -- Running IMP.isd making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/isd.h -- Found FFTW3 -- Performing Test FFTW3_found -- Performing Test FFTW3_found - Success -- Found FFTW3 OpenCV21 not found -- Found OPENCV22 -- Performing Test OpenCV22_found -- Performing Test OpenCV22_found - Success -- Found OpenCV22 Configuring module em2d ... IMP.domino not found Different data/build_info/sorted_modules -em2d Module IMP.em2d disabled ExampleDependency not found example is disabled Different data/build_info/sorted_modules -example Module IMP.example disabled -- Found FFTW3 -- Found FFTW3 multifit is disabled Different data/build_info/sorted_modules -multifit Module IMP.multifit disabled Configuring module pepdock ... IMP.domino not found Different data/build_info/sorted_modules -pepdock Module IMP.pepdock disabled Configuring module probC ... yes -- Running setup_swig_wrappers probC -- Running IMP.probC making all header -- Updating /Users/marco/IMP/build-fast/include/IMP/probC.h -- Found LIBTAU -- Performing Test libTau_found -- Performing Test libTau_found - Success -- Found libTau Configuring module cnmultifit ... IMP.multifit not found Different data/build_info/sorted_modules -cnmultifit Module IMP.cnmultifit disabled -- Running setup_swig_dependencies -- Creating thread pool with 8 Configuring application EMageFit ... no Different data/build_info/applications -EMageFit Application EMageFit disabled Configuring application cnmultifit_tools ... no Different data/build_info/applications -cnmultifit_tools Application cnmultifit_tools disabled Configuring application em_tools ... yes Configuring application example_app ... yes Configuring application foxs ... yes Configuring application integrative_docking ... no Different data/build_info/applications -integrative_docking Application integrative_docking disabled Configuring application ligand_score ... yes Configuring application multifit_tools ... no Different data/build_info/applications -multifit_tools Application multifit_tools disabled Configuring application pdb_tools ... yes Configuring application rmf_tools ... yes Configuring application saxs_merge ... yes Configuring application saxs_tools ... yes -- Running setup_imppy -- Documentation disabled as doxygen not found -- Configuring done -- Generating done -- Build files have been written to: /Users/marco/IMP/build-fast 5 - Then I launch from build-fast the command ninja -j1 RMF and I obtained the following error. [98/118] Swig source FAILED: cd /Users/marco/IMP/build-fast/src/dependency/RMF/swig && /usr/local/Cellar/cmake/3.1.0/bin/cmake -E make_directory /Users/marco/IMP/build-fast/lib && swig -python -castmode -naturalvar -Wextra -fvirtual -DPySwigIterator=RMF_PySwigIterator -c++ -DSwigPyIterator=RMF_SwigPyIterator -DRMFEXPORT -outdir /Users/marco/IMP/build-fast/lib -c++ -I/Users/marco/IMP/build-fast/include -I/usr/local/include -I/Users/marco/IMP/build-fast/src/dependency/RMF/src -I/Users/marco/IMP/src/modules/rmf/dependency/RMF_source/src -I/usr/local/Cellar/hdf5/1.8.14/include -I/Users/marco/IMP/build-fast/src/dependency/RMF/include -I/Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include -I/System/Library/Frameworks/Python.framework/Headers -I/Users/marco/IMP/src/swig -I/Users/marco/IMP/src/modules/rmf/dependency/RMF_source/swig -o /Users/marco/IMP/build-fast/lib/RMFPYTHON_wrap.cxx /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/swig/RMF.i /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments /Users/marco/IMP/src/modules/rmf/dependency/RMF_source/include/RMF/keys.h:41: Error: Macro 'RMF_DECLARE_KEY' expects 2 arguments ninja: build stopped: subcommand failed. In the mac I have the following versions of swig , clang , python and hdf5 SWIG Version 3.0.2 Apple LLVM version 6.0 (clang-600.0.56) (based on LLVM 3.5svn) Python 2.7.5 hdf5-1.8.14 I hope this message is enough to understand my problem. Best regards and thanks in advance for your help, Marco Di Stefano

3 2

missing binaries?
by Joel Mackay 16 Dec '14

16 Dec '14

Hi all I have just installed IMP 2.3 and am puzzled by the absence of binaries such as idock and emagefit. I couldn't see any indication in the changelog that they had been discontinued. I also get errors if I type "multifit help" or "cnmultifit help" (recommended commands in the tutorial). The error that comes up is: Traceback (most recent call last): File "/usr/bin/multifit", line 3, in <module> import IMP ImportError: No module named IMP I am wondering if I have done something wrong in the installation. Thanks Joel ------------------------------------------------------------------------------------------------------------------------------ Professor Joel Mackay | NHMRC Senior Research Fellow School of Molecular Bioscience THE UNIVERSITY OF SYDNEY Rm 674, Building G08 | The University of Sydney | NSW | 2006 | Australia T +61 2 9351 3906 | F +61 2 9351 5858 E joel.mackay(a)sydney.edu.au | W www.mmb.usyd.edu.au/mackay/

2 2

installation
by Joel Mackay 15 Dec '14

15 Dec '14

Hi I wanted to install IMP on an HPC system that is running the following: LSB_VERSION="core-2.0-noarch:core-3.2-noarch:core-4.0-noarch:core-2.0-x86_64:core-3.2-x86_64:core-4.0-x86_64" SGI Accelerate 1.4, Build 706r14.sles11sp2-1204092008 SGI Foundation Software 2.6, Build 706r14.sles11sp2-1204092008 SGI MPI 1.4, Build 706r14.sles11sp2-1204092008 SGI UPC 1.4, Build 706r14.sles11sp2-1204092008 SUSE Linux Enterprise Server 11 (x86_64) VERSION = 11 PATCHLEVEL = 2 Will I need to compile or might one of the binaries work? What do you recommend as the best strategy? Thanks Joel ------------------------------------------------------------------------------------------------------------------------------ Professor Joel Mackay | NHMRC Senior Research Fellow School of Molecular Bioscience THE UNIVERSITY OF SYDNEY Rm 674, Building G08 | The University of Sydney | NSW | 2006 | Australia T +61 2 9351 3906 | F +61 2 9351 5858 E joel.mackay(a)sydney.edu.au | W www.mmb.usyd.edu.au/mackay/

2 1

IMP 2.3.1 release
by Ben Webb 01 Dec '14

01 Dec '14

IMP 2.3.1 is now available. This is a bug fix and minor update release. Most notably two complete modeling examples that use the IMP::pmi module are now fully functional: modeling of the Nup84 subcomplex of the nuclear pore complex and of RNA Polymerase II. See https://github.com/salilab/imp/releases/tag/2.3.1 for the full change log. Check it out at http://integrativemodeling.org/ Ben -- ben(a)salilab.org http://salilab.org/~ben/ "It is a capital mistake to theorize before one has data." - Sir Arthur Conan Doyle

1 0

Re: [IMP-users] missing binaries?
by Joel Mackay 25 Nov '14

25 Nov '14

Hi Ben We are running redhat5 and have also tried redhat6 and fedora 20. It seems that the things that are missing change with each OS. Fedora20 was best. Redhat5 was missing idock and emagefit (among other things) and redhat6 was still missing idock. Is it because they RPM senses whether the OS will be able to handle each binary and decides whether or not to install it on that basis? We tried installing the RPM but also building from source. I think the other problem, relating to the missing IMP module, is due to conflicts with python versions - that is, we have several pythons on the system and the IMP modules were being built into a python that was not the default one that you get from the command line... Cheers Joel ------------------------------------------------------------------------------------------------------------------------------ Professor Joel Mackay | NHMRC Senior Research Fellow School of Molecular Bioscience THE UNIVERSITY OF SYDNEY Rm 674, Building G08 | The University of Sydney | NSW | 2006 | Australia T +61 2 9351 3906 | F +61 2 9351 5858 E joel.mackay(a)sydney.edu.au | W www.mmb.usyd.edu.au/mackay/

2 1

IMP 2.3.0 release
by Ben Webb 24 Oct '14

24 Oct '14

IMP 2.3.0 is now available, containing bug fixes and updates. Most notably the experimental IMP::pmi module is now available. This provides a variety of high-level Python classes to simplify the construction of a modeling protocol in IMP. One recent use of PMI is the modeling of the Nup84 subcomplex of the nuclear pore complex. We also now provide binary packages for Ubuntu LTS and 64-bit Windows. See https://github.com/salilab/imp/releases/tag/2.3.0 for the full change log. Check it out at http://integrativemodeling.org/ Ben -- ben(a)salilab.org http://salilab.org/~ben/ "It is a capital mistake to theorize before one has data." - Sir Arthur Conan Doyle

1 0

converting .rmf to .pdb
by Josh Bullock 13 Oct '14

13 Oct '14

Hello, I'm having trouble working out how to convert .rmf to .pdb files. It seems that pdb files can be written using the AnalysisReplicaExchange0 module, but is there a more straightforward method ? Many thanks, Josh

2 1

Configuring IMP 2.2.1: HDF5 once found, twice not
by Frank Thommen 03 Sep '14

03 Sep '14

Dear all, I am building IMP by using a HDF5 installation in a non-standard location. When running cmake with the HDF5_* options, HDF5 is found when configuring some modules, but not found when configuring others. I'm configuring IMP with: $ cmake ../imp-2.2.1 \ -DCMAKE_INSTALL_PREFIX=/my/imp/prefix \ -DBOOST_ROOT=/my/boost/root \ -DHDF5_LIBRARIES=/my/hdf5/root/lib \ -DHDF5_INCLUDE_DIRS=/my/hdf5/root/include And I get: Configuring module parallel ... yes -- Running setup_swig_wrappers parallel -- Running IMP.parallel making all header -- Updating /my/imp/prefix/SRC/build/include/IMP/parallel.h HDF5 not found and Configuring module saxs ... yes -- Running setup_swig_wrappers saxs -- Running IMP.saxs making all header -- Updating /my/imp/prefix/SRC/build/include/IMP/saxs.h HDF5 not found *but* I also get: Configuring module misc ... yes -- [...] -- Found HDF5: /my/hdf5/root/lib -- [...] I can overcome the problem by using the $CMAKE_INCLUDE_PATH and $CMAKE_LIBRARY_PATH environment variables instead. In this case HDF5 is found for all modules. Still, I think this behaviour is quite confusing (Problem already existed with IMP 2.2.0) Cheers Frank

2 2

Wrong CMake warning when building IMP 2.2.1
by Frank Thommen 03 Sep '14

03 Sep '14

Dear all, I am building IMP by using a Boost installation in a non-standard location. When running cmake with "-DBOOST_ROOT=/my/boost/root", the configuration ends successfully but at the end a warning is issued: ---------------- CMake Warning: The variable, 'BOOST_ROOT', specified manually, was not used during the generation. ---------------- However I don't think this warning is true, because if I run cmake w/o "-DBOOST_ROOT=/my/boost/root", I run into a configuration error: ---------------- CMake Error at /g/software/linux/pack/cmake-2.8.4/share/cmake-2.8/Modules/FindBoost.cmake:1128 (message): Unable to find the requested Boost libraries. Boost version: 1.33.1 Boost include path: /usr/include Detected version of Boost is too old. Requested version was 1.38 (or newer). The following Boost libraries could not be found: boost_system boost_filesystem boost_thread boost_program_options No Boost libraries were found. You may need to set BOOST_LIBRARYDIR to the directory containing Boost libraries or BOOST_ROOT to the location of Boost. etc. etc. ---------------- So in fact the variable BOOST_ROOT *is* used and the warning is wrong (or at least misleading). This warning didn't appear when configuring IMP 2.2.0 Cheers Frank

2 2

← Newer
1
...
12
13
14
15
16
17
18
...
28
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

IMP-users