loos-announce Mailing List for LOOS
Analyze molecular simulation data
Brought to you by:
alan_grossfield,
drtromo
You can subscribe to this list here.
| 2011 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(2) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2012 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
(1) |
Sep
(1) |
Oct
|
Nov
|
Dec
|
| 2013 |
Jan
|
Feb
(2) |
Mar
|
Apr
(1) |
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
(1) |
Oct
|
Nov
|
Dec
(2) |
| 2014 |
Jan
|
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(1) |
| 2015 |
Jan
(2) |
Feb
|
Mar
|
Apr
(1) |
May
(1) |
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
(2) |
| 2016 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
(1) |
| 2019 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: LOOS <loo...@gm...> - 2019-02-27 21:12:27
|
Hi LOOSers, We just released LOOS 3.0.0 on GitHub (https://github.com/GrossfieldLab/loos <https://github.com/GrossfieldLab/loos>). This is a big change, including a switch from python 2.7 to python 3.x, support for building in a Conda environment, a new reweighting facility for computing weighted averages in LOOS, and a bunch of small fixes and improvements as well. As part of this change, we’re also migrating away from Sourceforge as a platform. The Sourceforge page now redirects to github, so the only remnant of sourceforge is these mailing lists. Accordingly, we now recommend that the best way to request help, suggest a new feature, etc, is to raise an issue on Github. Also, if you want to be notified of new releases, etc, please follow LOOS on Github. For those of you attending the Biophysical Society meeting, please stop by our poster — it’s on Sunday, board B470. Sincerely, Alan and Tod ________________________________ Dr. Alan Grossfield Associate Professor Department of Biochemistry and Biophysics University of Rochester Medical Center 610 Elmwood Ave, Box 712 Rochester, NY 14642 Phone: 585 276 4193 http://membrane.urmc.rochester.edu https://orcid.org/0000-0002-5877-2789 |
|
From: LOOS <loo...@gm...> - 2016-12-05 18:13:28
|
Hi all, Just to follow up: only a few people have responded, and we _really_ need this information to help us set the direction of future LOOS development. The survey should take 5 minutes — please fill it out. Sincerely, Alan > On Nov 30, 2016, at 1:11 PM, LOOS <loo...@gm...> wrote: > > Hi everyone, > > We’re trying to get a better idea how people use LOOS, and what they do and don’t like about it. Accordingly, we’ve prepared a short survey: > > https://www.surveymonkey.com/r/RVGVLXW <https://www.surveymonkey.com/r/RVGVLXW> > > We’d really appreciate it you could take the survey, and pass the link to anyone else you know who uses LOOS (multiple responses from the same lab are fine with us). If you have additional comments, feel free to post them to this mailing list as replies to this message (or email them to us directly, if you’d prefer privacy). > > Thanks, > > Alan > > ___________________________________ > Dr. Alan Grossfield > Associate Professor > Department of Biochemistry and Biophysics > University of Rochester Medical Center > 610 Elmwood Ave, Box 712 > Rochester, NY 14642 > Phone: 585 276 4193 > http://membrane.urmc.rochester.edu <http://membrane.urmc.rochester.edu/> ___________________________________ Dr. Alan Grossfield Associate Professor Department of Biochemistry and Biophysics University of Rochester Medical Center 610 Elmwood Ave, Box 712 Rochester, NY 14642 Phone: 585 276 4193 http://membrane.urmc.rochester.edu |
|
From: LOOS <loo...@gm...> - 2016-11-30 18:11:11
|
Hi everyone, We’re trying to get a better idea how people use LOOS, and what they do and don’t like about it. Accordingly, we’ve prepared a short survey: https://www.surveymonkey.com/r/RVGVLXW <https://www.surveymonkey.com/r/RVGVLXW> We’d really appreciate it you could take the survey, and pass the link to anyone else you know who uses LOOS (multiple responses from the same lab are fine with us). If you have additional comments, feel free to post them to this mailing list as replies to this message (or email them to us directly, if you’d prefer privacy). Thanks, Alan ___________________________________ Dr. Alan Grossfield Associate Professor Department of Biochemistry and Biophysics University of Rochester Medical Center 610 Elmwood Ave, Box 712 Rochester, NY 14642 Phone: 585 276 4193 http://membrane.urmc.rochester.edu |
|
From: LOOS M. <loo...@gm...> - 2016-01-13 19:05:44
|
This release of LOOS is a bug-fix release. A bug was discovered in membrane_map that caused the z-axis of all coordinates to be set to 0. This affects height and vector calculations, but not density. The build system has been changed slightly. Some compilers require multiple environment variables to work correctly. In order to handle these cases, the LOOS build will now import all environment variables into SCons before building. This is *not* the SCons way to do things, however it makes handling these edge cases much easier. In the event that these extra environment variables cause problems, you can revert to a mostly "clean" build environment by editing the SConstruct file. Starting at line 72 with "env = Environment(...)", uncomment the first invocation and comment out the second one. In addition, improvements were made to the documentation. --Tod |
|
From: Dow H. <dp...@un...> - 2015-12-11 16:44:19
|
We use either linux mint 17 LTS or Ubuntu 14.04 LTS. Compilers used are usually the default gcc that comes with the distribution. Dow On Dec 11, 2015 10:03 AM, "LOOS" <loo...@gm...> wrote: > Hi all, > > As part of our efforts to keep our community happy, we need to know what > OS’s people are using LOOS on, and more importantly which OS’s they would > like to use LOOS on. We test a fair number of linux distributions (~25, > going back about 5 years) so that we can give people detailed instructions > for their native environment, but it’s important to us to make sure that > everything is as straightforward as possible. > > We would appreciate it if you would sent us the information about your OS > (e.g. Fedora 18 on x86_64), as well as anything unusual about it (e.g. > using a custom-installed version of the compiler). Please reply directly > to loo...@gm... rather than the list. > > Thanks, > > Alan and Tod > > ___________________________________ > Dr. Alan Grossfield > Associate Professor > Department of Biochemistry and Biophysics > University of Rochester Medical Center > 610 Elmwood Ave, Box 712 > Rochester, NY 14642 > Phone: 585 276 4193 > http://membrane.urmc.rochester.edu > > > > ------------------------------------------------------------------------------ > > _______________________________________________ > Loos-announce mailing list > Loo...@li... > https://lists.sourceforge.net/lists/listinfo/loos-announce > > |
|
From: LOOS <loo...@gm...> - 2015-12-11 15:03:24
|
Hi all, As part of our efforts to keep our community happy, we need to know what OS’s people are using LOOS on, and more importantly which OS’s they would like to use LOOS on. We test a fair number of linux distributions (~25, going back about 5 years) so that we can give people detailed instructions for their native environment, but it’s important to us to make sure that everything is as straightforward as possible. We would appreciate it if you would sent us the information about your OS (e.g. Fedora 18 on x86_64), as well as anything unusual about it (e.g. using a custom-installed version of the compiler). Please reply directly to loo...@gm... rather than the list. Thanks, Alan and Tod ___________________________________ Dr. Alan Grossfield Associate Professor Department of Biochemistry and Biophysics University of Rochester Medical Center 610 Elmwood Ave, Box 712 Rochester, NY 14642 Phone: 585 276 4193 http://membrane.urmc.rochester.edu |
|
From: LOOS M. <loo...@gm...> - 2015-11-23 20:55:13
|
Hot in time for the holidays, we are pleased to announce the release of LOOS 2.3.0. This release of LOOS comes with a number of major changes, most related to PyLOOS. A detailed description of these changes is available in the new "what's new in PyLOOS" section of the documentation, and is listed in the "Quick Links" off the main page. NUMPY IS NOW REQUIRED FOR BUILDING PYLOOS. For Linux users, this will just be a package install. For Mac users, the system numpy will be used. See the INSTALL file for more information. PyLOOS has been reorganized and new trajectory handling classes have been added. All Python-based components of PyLOOS are now in the loos.pyloos module. The new trajectory classes are easier to use than the old PyTraj classes (they no longer require you to explicitly wrap a loos.Trajectory object). In addition, you can combine multiple trajectories into a large "virtual" trajectory. The frames of the composite trajectory can be iteratively aligned (a la aligner) or aligned to a reference structure. See the documentation for pyloos under the Namespace tab or search for Trajectory. Additional PyLOOS changes include bug fixes for the iterativeAlignment function. The splitting functions in AtomicGroup now return a Python list rather than an AtomicGroupVector. Coordinates can be extracted from an AtomicGroup or a Trajectory as a NumPy matrix. An SVD/PCA function has also been added to PyLOOS. All old PyLOOS functions that begin or end with Py are now deprecated and will be removed in the next release. These include PyTraj, PyAlignedTraj, and iterativeAlignmentPy. Finally, a k-means clustering program (using PyLOOS) has been added. Also, we've added a new package (Voronoi), which will help users analyze membrane/membrane protein simulations. There are a number of significant non-PyLOOS changes as well. An "index" keyword has been added to pick atoms based on their position in a model file (rather than atom id). The rmsds tool (all-to-all RMSD) has been significantly optimized and can now be run multithreaded. The default is to use only a single thread. If you built LOOS with a multithreaded math library, be aware of possible conflicts (though that shouldn't happen with rmsds). Last, but not least, a number of minor bug fixes are included. See the ChangeLog file or the change log in the documentation for more details. *IMPORTANT NOTE FOR MACOS 10.11 "EL CAPITAN" USERS* There is a problem with using PyLOOS with the new System Integrity Protection (SIP) enabled (see https://support.apple.com/en-us/HT204899 for more information about SIP). We are aware of this and working on a decent solution. See the Mac section of the INSTALL file for more details. --The LOOS Developers |
|
From: Luke S. <luk...@gm...> - 2015-06-12 17:13:56
|
Good morning. I tried running "decorr_time" with an AMBER trajectory. I used this command: decorr_time --selection 'name == "CA"' --nrange 2,3,4 --frac 0.1 system.prmtop system.mdcrd >decorr.asc 10:10:250 I got this error message: terminate called after throwing an instance of 'loos::FileOpenError' what(): Error while opening 1i5n_0-250ns_prod.mdcrd Aborted (core dumped) Any tips? Lucas |
|
From: Tyler H. <har...@gm...> - 2015-05-12 18:47:34
|
After finishing the scon installation which seemed to complete no problem and finished with the phrase "scons: done building targets" I tried to source the setup.sh script both in my .bash_profile and in the command line. Both times, it seems to somehow break my shell and produces the following: $ source setup.sh-pre -bash: uname: command not found -bash: grep: command not found -bash: grep: command not found $ vi setup.sh-pre -bash: vi: command not found $ vim setup.sh-pre -bash: vim: command not found I also tried adding the !#/bin/sh or the !#/bin/bash to no avail. In case it is useful, I am using MacOS X 10.7 and did use the custom.py to install loos because netcdf and boost were installed using macports and are in the /opt/local/ directory. However, scons seems to have found everything and whatnot. Any idea what is causing this and how to rectify the problem? Thanks, Tyler |
|
From: LOOS M. <loo...@gm...> - 2015-04-27 12:21:52
|
Dear LOOSians, We are please to announce the release of version 2.2.5. This release of LOOS includes several important bug fixes and new functionality. Several issues involving Amber files have been addressed, including not correctly handling non-netCDF files when building with netCDF support. In addition, ".netcdf" is now a recognized file suffix for Amber trajectories. In PyLOOS only, a bug affecting operator overloads for GCoords has been fixed. Added functionality includes the ability to sort the input trajectory list numerically in merge-traj (the number used to sort is extracted from the filename using a regular expression). The default selection for SVD and ENM tools has changed to alpha-carbons. The tools "porcupine" and "enmovie" have been changed to make scaling the vectors and motion easier, along with selecting what modes to use. Note that the command line options and usage have changed and are not compatible with previous versions. For more details about these and other changes, see the ChangeLog or the documentation. |
|
From: LOOS M. <loo...@gm...> - 2015-01-26 21:51:29
|
Dear LOOSites,
This is another maintenance release of LOOS. It addresses a PyLOOS bug
that caused some values to be wrapped by SWIG that should not have been
(e.g. coord.x() would return a wrapped float rather than the float value).
This bug likely affects most PyLOOS programs and tools (including the
OptimalMembraneGenerator package). In addition, the error handling in the
build system has been improved. We strongly encourage all users to update
to this release.
|
|
From: LOOS M. <loo...@gm...> - 2015-01-18 01:22:22
|
Dear LOOSons, regrettably, a bug was found in the 2.2.0 release that prevented parsing Amber parmtop files. This new release fixes this box and includes improvements to the build system, along with support for Fedora 21 and a few other minor improvements. The build system changes were actually part of the 2.2.1 release. However, before an announcement was made about the release, the Amber bug was uncovered, so no official announcement of 2.2.1 was made. For more details about the changes to both 2.2.2 and 2.2.1, please see the README file, the ChangeLog, or the changes summary in the Documentation. And, as always, if you find any bugs or have any questions about LOOS, please contact us at loo...@gm.... We greatly appreciate the feedback and reports from our users! The LOOS Developers |
|
From: LOOS M. <loo...@gm...> - 2014-12-28 07:20:18
|
The LOOS Elves have been busy and are pleased to announce a new release of LOOS, version 2.2.0. This release of LOOS includes a number of significant improvements and changes. We are introducing the TrajectoryWriter class for writing trajectories. Two formats are currently supported: CHARMM/NAMD DCD format (DCDWriter) and Gromacs XTC (XTCWriter). Some tools now support picking the output trajectory format from the command line. Another major change is a complete overhaul of exceptions in LOOS and PyLOOS. For the tool-user, there should be no change other than some error messages may have a different format or wording. For Python users/developers, this change means that the Python interpreter will no longer crash when an exception occurs, and that you can use try/except in your Python code. Other changes to LOOS include a new "Optimal Membrane Generator" PyLOOS package, support for periodic boxes in Tinker XYZ and ARC files, and the ability to handle PDB files that are missing some fields (to the right of the coordinates). A backbone keyword has been added to the selection language. Finally, a bug affecting aligner when NOT using the entire trajectory has been fixed, along with a couple of other bugs. For more information about these changes, see the included Doxygen documentation or the code. --The LOOS Developers |
|
From: LOOS M. <loo...@gm...> - 2014-07-21 20:14:30
|
Dear LOOSites, we are pleased to announce the latest release of LOOS, version 2.1.3. This release addresses a couple of minor bugs, including a problem reading double-precision XTC files. Several new tools were added, including calculating molecular order parameters and a tool for looking at histograms of time-series (for convergence). In addition, several existing tools have new options and features. For more details, see the bundled ChangeLog file or the online documentation. --The LOOS Developers |
|
From: LOOS M. <loo...@gm...> - 2014-02-14 21:58:26
|
Dear LOOSies, we have just released version 2.1.2. This release addresses two bugs that crept into LOOS. One caused LOOS to be unable to read Amber mdcrd files, and the other caused the new PyTraj and PyAlignedTraj classes for PyLOOS to not be installed. The latter only affected those who install rather than use the build directory directly. A new feature has been added to xy_rdf: the --reselect option. This handles cases where molecules move back and forth between leaflets. As always, if you run into any problems, please contact us. --The LOOS Development Team |
|
From: LOOS M. <loo...@gm...> - 2014-02-02 03:02:32
|
Dear LOOSites, we are please to announce the next release of LOOS, version 2.1.1. This release includes a few bug fixes, addresses an inconsistency in the build script for the examples, and sports a few important improvements to PyLOOS. In addition, there is a new tool (packing_score) for quantifying the packing between selections over the course of a trajectory. In PyLOOS, some sequence-like objects (e.g. GCoord, AtomicGroup, TimeSeries, and Matrix44) were not iterable Python objects and trying to use them as such would fill memory. These objects will now raise a Python exception and can be used as an iterable object. In addition, we have added two new Python-only trajectory classes: PyTraj and PyAlignedTraj. These wrap a LOOS Trajectory (and AtomicGroup) and make it into an iterable Python object. You can also specify a skip and a stride through the trajectory (or provide your own iterator for selecting frames). See the class documentation for more details, or look in Packages/PyLOOS for examples. We have also addressed an issue with building PyLOOS in recent versions of MacOS 10.9 "Mavericks". There is an apparent problem with how STL containers are wrapped. For now, we have disabled the wrapping of iterator methods for STL containers, so member functions such as erase() and begin() will not be available in PyLOOS on this platform. If you run into any problems with this release, please don't hesitate to contact us! --The LOOS Development Team |
|
From: LOOS M. <loo...@gm...> - 2013-12-12 20:48:36
|
Dear LOOSers, I hope you all are enjoying our latest release, version 2.10. We here at LOOS Development Central thought you might want to know we've created 2 new mailing lists, one focused on issues involved in using LOOS, the other on development. If you are interested in subscribing, you can do so here: https://lists.sourceforge.net/lists/listinfo/loos-users https://lists.sourceforge.net/lists/listinfo/loos-devel Cheers, The LOOS Development Team (Alan & Tod) |
|
From: LOOS M. <loo...@gm...> - 2013-12-06 21:17:34
|
Dear LOOSers, we are pleased to announce a new release of LOOS. This release contains several major changes to LOOS. We have rewritten the build system so that it will configure LOOS based on your system (i.e. using the Autoconf functionality of SCons) rather than use pre-set build configurations for common distributions. We are now deprecating a few older Linux distributions, however we have added support for MacOS 10.9 (Mavericks). Along with this, LOOS now requires that you have at least Boost version 1.36 installed. This will require manually installing a more recent Boost for some older Linux distributions. We are also introducing a major change in how Trajectory classes work, including changes to the Atom and AtomicGroup classes. Briefly, LOOS now uses an atom index, assigned when a model is read, for updating from a trajectory rather than the atomid's. Also included in this release are a few minor bug fixes and a new tool, membrane_map, which computes the 2D distribution of various physical properties around a membrane protein For more details, please see the README and INSTALL files, as well as the "Building LOOS" section of the documentation ( http://loos.sourceforge.net/building.html ) and the change log ( http://loos.sourceforge.net/_changes.html#release210 ). If you run into any problems with this release, please don't hesitate to contact us! ---The LOOS Development Team |
|
From: LOOS M. <loo...@gm...> - 2013-09-13 15:15:05
|
Dear LOOSers, We are pleased to announce the release of LOOS version 2.0.6. This release addresses a couple of bugs in gmxdump2pdb.pl and serialize-selection. We have added a tool (fixdcd) to correct the DCD header for trajectories that think they have no frames or an incorrect number of frames. Also included are a number of minor bug fixes and better error-handling in some of the tools. Improvements include disambiguating the direction of PCA vectors in cross-dist, a smoothed contact matrix for transition_contacts, and a new PyLOOS tool, "protein_tilt.py". In addition, the python tools are now installed, if you are building the PyLOOS version of LOOS. For more detailed information about the changes in this release, see either the included README or ChangeLog file. Just a reminder again that we have a new contact email for help with LOOS: loo...@gm... Let the downloading begin... Tod, Alan, and all the local LOOSers |
|
From: LOOS M. <loo...@gm...> - 2013-06-21 20:46:24
|
Dear LOOSers, We are pleased to announce the release of LOOS version 2.0.5. There are a number of internal changes in this version, including some refactoring inside various Trajectory classes and subclasses, a new HBondDetector class to let you easily check for hydrogen bonds using distance and angle criteria, and changing how some tools cache structures. PyLOOS has been modified to handle LOOS exceptions, so that if LOOS throws it won't crash the python interpreter. In addition, renum-pdb has changed so that it correctly retains connectivity information if present in the original, and aligner has a new option to only align in the xy plane. The native_contacts tool has been updated to be more consistent with the new transition_contacts tool (which the difference in contacts between 2 structures as a reaction coordinate for analyzing a protein simulation. There's also a new serialize-selection tool, which takes a trajectory containing multiple copies of a molecule (e.g. a peptide) and produces a new trajectory where each frame contains just a single copy of the peptide. Finally, there's a new fcontacts tool that should, in most cases, supercede the contact-time tool. It calculates the contacts made between a probe and a set of target selections as a fraction of all contacts with the probe. Finally, you may notice that there's a new contact email for help with LOOS: loo...@gm... All in all, a worthy release. Let the downloading begin... Alan, Tod, and all the local LOOSers |
|
From: Loos M. <loo...@ur...> - 2013-04-05 16:27:46
|
We've just released the 2.0.4 version of LOOS. This release primarily addresses an issue when building LOOS with a recent version of Boost (1.52 or later). There are also several changes to the PyLOOS beta: we now support MacOS and you may also install PyLOOS along with LOOS. This required a change in how PyLOOS is built however. Rather than using "scons pyloos", use "scons pyloos=1". For more detailed information, please see either the ChangeLog or INSTALL file, or consult the Doxygen-generated documentation. |
|
From: Loos M. <loo...@ur...> - 2013-02-11 20:02:33
|
Dear LOOS community, in the push to release a new LOOS for the Biophysical Society meeting, a binary version was inadvertently released instead of the official source code release. If you downloaded the 2.0.3 tar file, please download the 2.0.3b tar file and build this version. We sincerely apologize for any inconvenience and confusion this errant release may have caused. The LOOS Maintenance Team |
|
From: LOOS <loo...@ur...> - 2013-02-07 15:30:10
|
This message was supposed to go out last week, but seems to have gotten hung up; if you've downloaded since 2/2, you have the most recent version. Dear LOOSers, We are pleased to announce the release of LOOS 2.0.3, just in time for the Biophysical Society Meeting. This release includes a number of bug fixes and new features, including support for the AMOEBA variant of Amber prmtop files as well as NetCDF formatted Amber trajectories (if a suitable NetCDF library is installed prior to building LOOS). The gridgauss tool has been changed to allow more control over the smoothing kernel. The command line arguments have therefore changed and the old way of running it will no longer work. The LOOS Maintenance Team ___________________________________ Dr. Alan Grossfield Assistant Professor Department of Biochemistry and Biophysics University of Rochester Medical Center 610 Elmwood Ave, Box 712 Rochester, NY 14642 Phone: 585 276 4193 http://membrane.urmc.rochester.edu |
|
From: LOOS <loo...@ur...> - 2012-09-06 19:53:06
|
Recently, we released version 2.0.2. This is a maintenance release, primarily to fix a bug in the handling of connectivity when reading PDB files. However, we did make a change to the model-select tool, making the selection option required, and adding a "splitby" option; "splitby" will let you test the effects of different splitting modes (residue, segid, molecule, atom name) on the selected subset of the system. Sincerely, The LOOS Development Team ___________________________________ Dr. Alan Grossfield Assistant Professor Department of Biochemistry and Biophysics University of Rochester Medical Center 610 Elmwood Ave, Box 712 Rochester, NY 14642 Phone: 585 276 4193 http://membrane.urmc.rochester.edu |
|
From: LOOS <loo...@ur...> - 2012-08-07 14:40:38
|
Dear LOOS users, Yesterday, we released version 2.0.1 of LOOS. This is primarily a maintenance release. There are a couple of bug fixes to aligner, xy_rdf, and a couple of water tools. In addition, there are a couple of new tools in the hydrogen bonds package. For developers, there was a significant change to how the Atom::bondsbit is handled. We also include a new perl script, gmxdump2pdb.pl, that will take the output from gmxdump and create pdb and psf files containing connectivity information. This is particularly valuable for visualizing coarse-grained simulations using MARTINI, since using the resulting psf will allow VMD to render those trajectories with correct bonding information (pymol can use either the PDB or the PSF). As always, we recommend that people upgrade promptly, and hope that anyone running into trouble will contact us with questions. Thanks, The LOOS Development Team ___________________________________ Dr. Alan Grossfield Assistant Professor Department of Biochemistry and Biophysics University of Rochester Medical Center 610 Elmwood Ave, Box 712 Rochester, NY 14642 Phone: 585 276 4193 http://membrane.urmc.rochester.edu ___________________________________ Dr. Alan Grossfield Assistant Professor Department of Biochemistry and Biophysics University of Rochester Medical Center 610 Elmwood Ave, Box 712 Rochester, NY 14642 Phone: 585 276 4193 http://membrane.urmc.rochester.edu |