Different X-Axis Values, But The Same X-Axis Units – Getting Excel 2013 (OSX-Specific) To Produce Multiple Scatter Plots On The Same Graph

Posting a workaround to re-introduce a feature for Excel 2013 that I think was removed for some reason and for which information on Excel 2013 (OSX-specific) is impossible to find through google searches. It is my hope that newer versions of Excel don’t have this thoroughly annoying problem (and if there’s an obvious way I don’t know about to make this happen in a single shot, feel free to drop a line).

If you found this page via google, I’m going to assume you were searching for something like the following questions (which I’m including below so that search engines find similar questions):

How do I add:
– data with two different X-axes to a single plot in Excel?
– multiple plots to the same graph in Excel with different X-values but the same X-axis units?
– a second dataset to a plot in Excel with new X-axis values?
– a new dataset with different abscissa values in Excel?
– a secondary X-axis to plot new data on the same graph in Excel?

My Scenario – Overlaying Two Spectra On The Same Graph

This issue came up for me when trying to generate some simple spectral overlays in Excel. The problem proceeds as follows:

1. You’ve two datasets with the same X-axis values.

2. You’ve a third dataset for which the X-axis points have the same units, but different values.

3. You plot this third dataset with Chart… -> Add Data by selecting your X-axis and Y-axis values (just selecting the columns).

4. You don’t get the expected results.

Ideally, Excel would see that the headers for the X-axis columns have the same exact labels in all the datasets and treat the new points (in Step 3) as values to be accounted for within the same range of numbers as the previous plot. If you plot dataset 1 and add dataset 2, there’s no problem (because the X-axis column is identical). Step 3 will set you back an hour or so (assuming this isn’t in the Help Pages, which take too long to load anyway).

The steps below describe a way to recover an old functionality in Excel 2007 that just simply worked without issue.

1. My Working Excel Sheet

To set the stage, my sheet is set up as shown below.

2014nov18_excel_1

Rows 7-3115 and 3122-3498 are hidden for clarity
Col A: cm-1 – my X-axis values, with label
Col B: Exp.: Time=0 – My first dataset
Col C: Exp.: Time=20 – My second dataset; uses the same X-axis values as Col B (cm-1)
Col F: cm-1 – my X-axis values for a new dataset (in G,H), scalable using Cell J2
Col G: Theory – my calculated values
Col H: Theory (Scaled) – my calculated values in Col G, scalable using Cell J5
Col J Rows 1,2: X Scaling – used to scale the values of the calculated data along the X-axis (actual values in Col E)
Col J Rows 4,5: Y Scaling – used to scale the values of the calculated data along the Y-axis (Col H)

To explain briefly (for the non-spectroscopists), your calculated X-axis (energy, here in wavenumbers) and Y-axis (intensity) values are/can be adjusted for better agreement with experiment. In this case, I’m applying a global scalar to the X-axis (Cell J2 – make the plot wider or narrower) and Y-axis (Cel J5 – make the plot taller or shorter) values. These values will be adjusted to taste in the final images. I’m going into a little more detail than one would otherwise need because how you make your data fit properly will depend on how you change the source data (and not doing it properly will lead to a fitting problem. You’ll see shortly).

2. First + Second Plottings

After plotting the first dataset (Col A + Col B) as a Scatter Plot and doing some cleaning up, you get the image below.

2014nov18_excel_2

To add the data in Col C (which we see already uses the same X-axis values as Col B), you simply go to Chart -> Add Data, select Col C, and you should get the image below (20 min data in green):

2014nov18_excel_3

3. Where The Trouble Starts

To this new plot we want to add the Theory results. You’ll note that the experimental X-axis values differ in their increments from the theoretical data. But X-axis values are X-axis values, right?

Go to Chart -> Add Data, then select the new X/Y data you want to include in the plot. When adding the data, you should see the following.

2014nov18_excel_4

Not cool. Instead of reading Col F as new X-axis values with the same units (with Excel 2007 did just fine), Excel 2013 sees this as a new dataset using the original X-axis values in Col A. The Theory plot in Col H (red) looks like it worked, but you’ll see shortly that the defaulting to the X-axis values in Col A has resulted in very poor agreement with experiment (and it’s the wrong X-axis values, so this is no surprise).

First off, the cm-1 (black) values are bogus, so delete that dataset to beautify the plot. That image is below.

2014nov18_excel_5

4. Partial Fix (Nearly The Same As The Full Fix)

We clean up the spectrum by right-clicking on the plot and choosing Select Data… (shown below)

2014nov18_excel_6

This will take you to a Select Data Source window where one of the Series will be selected. Click on the dataset you want to change the X-axis column for (here, Theory (Scaled)). You will see next to the obvious red arrow that the X values: column is reading $A$2:$A$3501. You’ll note that Col A has no data in cell A3501 (that’s bad enough).

2014nov18_excel_7

Now, for the Partial Fix, double-click on the content of X values: to select it all, and then click on the column you want the X-axis values to be (for me, Col F). X values: will become the following:

2014nov18_excel_8

If you do this and hit OK, you’ll see the plot below, which is just what you expected from a well-behaved Excel program.

2014nov18_excel_9

Your X-axis is as it should be, even if the peak intensities for the theory are too high. That you can remedy by changing Cell J5 (results below).

2014nov18_excel_10

That is much better, but you can see that the most prominent peaks (around 1600 cm-1) are calculated too high. This is why there’s a scaling factor in Cell J2. If you change the value of Cell J2 to, say, 0.973, you produce the following plot:

2014nov18_excel_10

Which, as you can see, is exactly the same as the previous plot. Our X-axis scaling factor had no affect.

5. And Now, The Full Fix

We selected the new X-axis column (Col F) correctly, but Excel won’t give us our proper scaling unless we specifically define the range of cells used for the X-axis values. So, we go back to Select Data… (right click) and put the actual cell numbers in. At the obvious red arrow below…

2014nov18_excel_12

Change the $F:$F to our actual range, $F$2:$F$3501

2014nov18_excel_13

Hit OK. If you didn’t change your J2 value back to 1, you should see that your plot slid right into place (granted, the theory doesn’t line up all that well anyway, but that’s a problem for a different post).

2014nov18_excel_14

And that’s it. May this post spare you the time wasted searching for a solution to a problem that didn’t previously exist.

NAMOT Pre-Release 2.2.0-pre4 In OSX 10.8 (Maybe Older Versions)

A recent visit to the College of Nanoscale Science and Engineering (CNSE) at SUNY Albany inspired a few new DNA ideas that I decided would be greatly simplified by having NAMOT available again for design. Having failed at the base install of the NAMOT 2 version and, unfortunately, not having NAMOT available in Fink for a simple installation, the solution became to build the pre-release from scratch. Ignoring the many errors one encounters while walking through an OSX/Xcode/Fink/X11 bootstrap, the final procedure worked well and without major problem. As usual, the error messages at varied steps are provided below because, I assume, those messages are what you’re searching for when you find your way here.

0. Required Installations

You’ll need the following installed for this particular build. I believe XCode is the only thing that you’ll have to pay for (if you don’t already have it. I seem to remember paying $5 through the App Store).

1. XCode

The OSX Developer Suite – developer.apple.com/xcode

2. XQuartz

An OSX (X.Org) X Window System – xquartz.macosforge.org/landing/

3. Fink

An OSX port program for a host of Unix codes and libraries – www.finkproject.org

3a. GSL

The GNU Scientific (C and C++) Libraries – www.gnu.org/software/gsl. This will be installed with Fink.

3b. LessTif

An OSF/Motif clone (made available for OSX through Fink) – lesstif.sourceforge.net. This will be installed with Fink.

4. NAMOT2.2.0-pre4

The -pre4 is currently available (from 2003) from sourceforge.net/projects/namot/files/. I did not try -pre3 and had no luck with the official 2 release.

And, with that…

1. XCode

Blindly follow the install procedure. Several steps below deal with working around the default install locations (specifically, /sw).

2. XQuartz

If you don’t have XQuartz installed, you’re configure step…

cd Downloads
cd namot-2.2.0-pre4
./configure 

will produce the following error…

checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for gawk... no
checking for mawk... no
checking for nawk... no
checking for awk... awk
...
creating libtool
checking for X... no
checking for main in -lX11... no
NAMOT requires Xwindows

Blindly follow the XQuartz install process. After the installations, you’ll receive the same error as above. The –x-libraries= and –x-includes= additions to configure below direct the script to the proper libraries and includes.

./configure --x-libraries=/usr/X11/lib/ --x-includes=/usrX11/include/

Hopefully, you’ll find yourself past the first install problem and onto the second problem.

checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for gawk... no
checking for mawk... no
checking for nawk... no
checking for awk... awk
...
creating libtool
checking for X... libraries /usr/X11/lib/, headers /usrX11/include/
checking for gethostbyname... yes
checking for connect... yes
checking for remove... yes
checking for shmat... yes
checking for IceConnectionNumber in -lICE... yes
checking for main in -lX11... yes
checking for main in -lgslcblas... no
NAMOT requires GNU Scientific Library

3. Fink

The next two codes that need to be installed are the GNU Scientific Libraries and LessTif, both of which are much easier to install using Fink. It is generally useful for many other codes as well, so a good program for any computational chemist to have on hand. The install should be non-problematic despite having to build it from source in 10.6 – 10.8 (as of January 2013). If you build with all the default settings, you’ll have no trouble after.

cd Downloads
cd fink-0.34.4
./bootstrap 

I chose the default settings throughout.

Fink must be installed and run with superuser (root) privileges. Fink can automatically try to become root when it's run from a user account. Since you're currently running this script as a normal user, the method you choose will also be used immediately for this script. Available methods:

(1)	Use sudo
(2)	Use su
(3)	None, fink must be run as root

Choose a method: [1] 

...

You should now have a working Fink installation in '/sw'. You still need package descriptions if you want to compile packages yourself. You can get them by running either of the commands: 'fink selfupdate-rsync', to update via rsync (generally preferred); or 'fink selfupdate-cvs', to update via CVS (more likely to work through a firewall).

Run '. /sw/bin/init.sh' to set up this terminal session environment to use Fink. To make the software installed by Fink available in all of your future terminal shells, add '. /sw/bin/init.sh' to the init script '.profile' or '.bash_profile' in your home directory. The program /sw/bin/pathsetup.sh can help with this. Enjoy.

Then you run the final step in Fink below:

/sw/bin/pathsetup.sh

Which will produce the following two pop-ups notifying you of shell modifications.

3a. GSL

With the install of Fink, you need to install GSL and LessTif. If you try to install either immediately after installation…

fink install gsl

…you’ll receive the following error:

Password:
Scanning package description files..........
Information about 305 packages read in 0 seconds.
no package found for "gsl"
Failed: no package found for specification 'gsl'!

Required after the installation is a fink selfupdate.

fink selfupdate

As usual, follow the default settings…

fink needs you to choose a SelfUpdateMethod.

(1)	cvs
(2)	Stick to point releases
(3)	rsync

Choose an update method [3] 
/usr/bin/find /sw/fink -name CVS -type d -print0 | xargs -0 /bin/rm -rf
fink is setting your default update method to rsync
...
Updating the list of locally available binary packages.
Scanning dists/stable/main/binary-darwin-i386
New package: dists/stable/main/binary-darwin-i386/base/base-files_1.9.13-1_darwin-i386.deb
New package: dists/stable/main/binary-darwin-i386/base/fink-mirrors_0.34.4.1-1_darwin-i386.deb

Which then leads to a successful GSL install.

fink install gsl

Producing the following output…

Information about 12051 packages read in 1 seconds.
The package 'gsl' will be built and installed.
Reading build dependency for gsl-1.15-1...
Reading dependency for gsl-1.15-1...
Reading runtime dependency for gsl-1.15-1...
Reading dependency for gsl-shlibs-1.15-1...
...
Updating the list of locally available binary packages.
Scanning dists/stable/main/binary-darwin-i386
New package: dists/stable/main/binary-darwin-i386/sci/gsl-shlibs_1.15-1_darwin-i386.deb
New package: dists/stable/main/binary-darwin-i386/sci/gsl_1.15-1_darwin-i386.deb

Attempting a fresh build after the GSL step…

./configure --x-libraries=/usr/X11/lib/ --x-includes=/usrX11/include/

…then still produces the following error:

checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for gawk... no
checking for mawk... no
checking for nawk... no
checking for awk... awk
...
checking for IceConnectionNumber in -lICE... yes
checking for main in -lX11... yes
checking for main in -lgslcblas... no
NAMOT requires GNU Scientific Library

As mentioned above, there are a few redirects that need to be made after the XCode / Fink install to put libraries and includes where, in this case, NAMOT expects them. To perform this task, we’ll be using symbolic links.

sudo ln -s /sw/include/gsl /usr/include/
sudo ln -s /sw/lib/libgsl* /usr/lib

Now attempting a build…

./configure --x-libraries=/usr/X11/lib/ --x-includes=/usrX11/include

Gets you past the GSL issue.

checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for gawk... no
checking for mawk... no
checking for nawk... no
checking for awk... awk
…
checking for IceConnectionNumber in -lICE... yes
checking for main in -lX11... yes
checking for main in -lgslcblas... yes
checking for main in -lgsl... yes
checking for XShmCreateImage in -lXext... yes
checking for main in -lXt... yes
checking for main in -lXm... no
NAMOT requires Motif...try LessTif(http://www.lesstif.org)

3b. LessTif

The LessTif symbolic links work the same as the GSL symbolic links. This fink install may take a while.

fink install lesstif

Output below…

Information about 12051 packages read in 1 seconds.
The package 'lesstif' will be built and installed.
Reading build dependency for lesstif-0.95.2-4...
Reading dependency for lesstif-0.95.2-4...
Reading runtime dependency for lesstif-0.95.2-4...
...
Setting up lesstif (0.95.2-4) ...
Clearing dependency_libs of .la files being installed

Updating the list of locally available binary packages.
Scanning dists/stable/main/binary-darwin-i386
New package: dists/stable/main/binary-darwin-i386/x11/app-defaults_20010814-12_darwin-i386.deb
New package: dists/stable/main/binary-darwin-i386/x11/lesstif-bin_0.95.2-4_darwin-i386.deb
New package: dists/stable/main/binary-darwin-i386/x11/lesstif-shlibs_0.95.2-4_darwin-i386.deb
New package: dists/stable/main/binary-darwin-i386/x11/lesstif_0.95.2-4_darwin-i386.deb
./configure --x-libraries=/usr/X11/lib/ --x-includes=/usrX11/include/

But, unfortunately, the LessTif libraries are not in the expected locations.

checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for gawk... no
checking for mawk... no
checking for nawk... no
checking for awk... awk
...
checking for IceConnectionNumber in -lICE... yes
checking for main in -lX11... yes
checking for main in -lgslcblas... yes
checking for main in -lgsl... yes
checking for XShmCreateImage in -lXext... yes
checking for main in -lXt... yes
checking for main in -lXm... no
NAMOT requires Motif...try LessTif(http://www.lesstif.org)

So we add the symbolic links…

sudo ln -s /sw/lib/libXm.* /usr/lib
sudo ln -s /sw/include/Xm /usr/include

Which, finally, runs configure…

./configure --x-libraries=/usr/X11/lib/ --x-includes=/usrX11/include/

…with no errors.

checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for gawk... no
checking for mawk... no
checking for nawk... no
checking for awk... awk
...
config.status: creating docs/demos/curve/Makefile
config.status: creating docs/demos/dit/Makefile
config.status: creating docs/demos/Makefile
config.status: creating config.h
config.status: executing depfiles commands

NOTE: The make step with Python 2.6 produces the following error below. I did not diagnose this beyond the failure to build under 10.6. OSX 10.8 comes with Python 2.7, which did not produce this problem (I’m assuming this is the origin of the problem).

make

…will produce the following error at the pngwriter.c step.

/bin/sh ../libtool --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I..    -DLIB_HOME="\"/usr/local/share/namot\"" -DHELP_FILE_DIR="\"/usr/local/share/namot\"" -I/System/Library/Frameworks/Python.framework/Versions/2.6/include/python2.6 -I/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/config -g -O2 -c -o _pynamot_la-pngwriter.lo `test -f 'pngwriter.c' || echo './'`pngwriter.c
gcc -DHAVE_CONFIG_H -I. -I. -I.. -DLIB_HOME=\"/usr/local/share/namot\" -DHELP_FILE_DIR=\"/usr/local/share/namot\" -I/System/Library/Frameworks/Python.framework/Versions/2.6/include/python2.6 -I/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/config -g -O2 -c pngwriter.c -MT _pynamot_la-pngwriter.lo -MD -MP -MF .deps/_pynamot_la-pngwriter.TPlo  -fno-common -DPIC -o _pynamot_la-pngwriter.lo
pngwriter.c: In function 'dump_PNG':
pngwriter.c:28: error: 'png_structp' undeclared (first use in this function)
pngwriter.c:28: error: (Each undeclared identifier is reported only once
pngwriter.c:28: error: for each function it appears in.)
pngwriter.c:28: error: expected ';' before 'png_ptr'
pngwriter.c:29: error: 'png_infop' undeclared (first use in this function)
pngwriter.c:29: error: expected ';' before 'info_ptr'
pngwriter.c:30: error: 'png_byte' undeclared (first use in this function)
pngwriter.c:30: error: 'row_pointers' undeclared (first use in this function)
pngwriter.c:30: error: expected expression before ')' token
pngwriter.c:31: error: 'png_text' undeclared (first use in this function)
pngwriter.c:31: error: expected ';' before 'text_ptr'
pngwriter.c:39: warning: incompatible implicit declaration of built-in function 'memset'
pngwriter.c:39: error: 'text_ptr' undeclared (first use in this function)
pngwriter.c:47: error: 'png_ptr' undeclared (first use in this function)
pngwriter.c:47: error: 'PNG_LIBPNG_VER_STRING' undeclared (first use in this function)
pngwriter.c:48: error: 'png_voidp' undeclared (first use in this function)
pngwriter.c:57: error: 'info_ptr' undeclared (first use in this function)
pngwriter.c:60: error: 'png_infopp' undeclared (first use in this function)
pngwriter.c:82: error: 'PNG_COLOR_TYPE_RGB' undeclared (first use in this function)
pngwriter.c:82: error: 'PNG_INTERLACE_ADAM7' undeclared (first use in this function)
pngwriter.c:83: error: 'PNG_COMPRESSION_TYPE_DEFAULT' undeclared (first use in this function)
pngwriter.c:83: error: 'PNG_FILTER_TYPE_DEFAULT' undeclared (first use in this function)
pngwriter.c:85: error: 'PNG_sRGB_INTENT_ABSOLUTE' undeclared (first use in this function)
pngwriter.c:90: error: 'PNG_TEXT_COMPRESSION_NONE' undeclared (first use in this function)
pngwriter.c:93: error: 'PNG_TEXT_COMPRESSION_zTXt' undeclared (first use in this function)
pngwriter.c:104: error: expected expression before ')' token
make[2]: *** [_pynamot_la-pngwriter.lo] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2

The build on 10.8 continues as below, with a few warnings about the symbolic link usage that do not seem to affect the program usability (or continued build).

make

Results below…

make  all-recursive
Making all in src
source='namot_wrap.c' object='_pynamot_la-namot_wrap.lo' libtool=yes \
	depfile='.deps/_pynamot_la-namot_wrap.Plo' tmpdepfile='.deps/_pynamot_la-namot_wrap.TPlo' \
	depmode=gcc3 /bin/sh ../depcomp \
	/bin/sh ../libtool --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I..    -DLIB_HOME="\"/usr/local/share/namot\"" -DHELP_FILE_DIR="\"/usr/local/share/namot\"" -I/System/Library/Frameworks/Python.framework/Versions/2.7/include/python2.7 -I/System/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/config -g -O2 -c -o _pynamot_la-namot_wrap.lo `test -f 'namot_wrap.c' || echo './'`namot_wrap.c

...

*** Warning: linker path does not have real file for library -lXm.
*** I have the capability to make that library automatically link in when
*** you link to this library.  But I can only do this if you have a
*** shared version of the library, which you do not appear to have
*** because I did check the linker path looking for a file starting
*** with libXm and none of the candidates passed a file format test
*** using a file magic. Last file checked: /sw/lib/libXm.la

*** Warning: linker path does not have real file for library -lgsl.
*** I have the capability to make that library automatically link in when
*** you link to this library.  But I can only do this if you have a
*** shared version of the library, which you do not appear to have
*** because I did check the linker path looking for a file starting
*** with libgsl and none of the candidates passed a file format test
*** using a file magic. Last file checked: /sw/lib/libgsl.la

*** Warning: linker path does not have real file for library -lgslcblas.
*** I have the capability to make that library automatically link in when
*** you link to this library.  But I can only do this if you have a
*** shared version of the library, which you do not appear to have
*** because I did check the linker path looking for a file starting
*** with libgslcblas and none of the candidates passed a file format test
*** using a file magic. Last file checked: /sw/lib/libgslcblas.la

*** Warning: libtool could not satisfy all declared inter-library
*** dependencies of module _pynamot.  Therefore, libtool will create
*** a static module, that should work as long as the dlopening
*** application is linked with the -dlopen flag.

...

Making all in libs
make[2]: Nothing to be done for `all'.
Making all in docs
Making all in helpfiles
make[3]: Nothing to be done for `all'.
Making all in demos
Making all in 6way
make[4]: Nothing to be done for `all'.
Making all in bending
make[4]: Nothing to be done for `all'.
Making all in cube
make[4]: Nothing to be done for `all'.
Making all in curve
make[4]: Nothing to be done for `all'.
Making all in dit
make[4]: Nothing to be done for `all'.
make[4]: Nothing to be done for `all-am'.
make[3]: Nothing to be done for `all-am'.
Making all in etc
make[2]: Nothing to be done for `all'.

Finally, the install…

make install

Which produces the following:

Making install in src
/bin/sh ../mkinstalldirs /usr/local/lib
 /bin/sh ../libtool --mode=install /usr/bin/install -c  _pynamot.la /usr/local/lib/_pynamot.la
/usr/bin/install -c .libs/_pynamot.lai /usr/local/lib/_pynamot.la
/usr/bin/install -c .libs/_pynamot.a /usr/local/lib/_pynamot.a
ranlib /usr/local/lib/_pynamot.a
chmod 644 /usr/local/lib/_pynamot.a
----------------------------------------------------------------------
Libraries have been installed in:
   /usr/local/lib

If you ever happen to want to link against installed libraries
in a given directory, LIBDIR, you must either use libtool, and
specify the full pathname of the library, or use the `-LLIBDIR'
flag during linking and do at least one of the following:
   - add LIBDIR to the `DYLD_LIBRARY_PATH' environment variable
     during execution

See any operating system documentation about shared libraries for
more information, such as the ld(1) and ld.so(8) manual pages.
...
/bin/sh ../mkinstalldirs /usr/local/share/namot
 /usr/bin/install -c -m 644 Namot2.512 /usr/local/share/namot/Namot2.512
 /usr/bin/install -c -m 644 Namot2.600 /usr/local/share/namot/Namot2.600
 /usr/bin/install -c -m 644 Namot2.700 /usr/local/share/namot/Namot2.700
 /usr/bin/install -c -m 644 icon1.xv /usr/local/share/namot/icon1.xv
make[2]: Nothing to be done for `install-exec-am'.
make[2]: Nothing to be done for `install-data-am'.

With luck, your launching of NAMOT will open XQuartz and produce a fully operational NAMOT session.

namot

And, for more assistance with producing DNA files for GROMACS, consider the Modifications To The ffG53a6.rtp And ffG53a5.rtp Residue Topology Files Required For Using GROMOS96-NAMOT-GROMACS v1, sed-Based Script For Converting NAMOT And NAMOT2 DNA Output To GROMOS96 Format For GROMACS Topology Generation v1, and sed-Based Script For Converting NAMOT And NAMOT2 DNA Output To ffAMBER Format For GROMACS Topology Generation v1 pages on this blog.

GROMACS 4.5.5, OpenMPI 1.6, And FFTW 3.3.2 Compilation Under Mountain Lion (OSX 10.8) With XCode (And A Little Help From Homebrew)

Minus a few glitches easily fixed with the right software, this build wasn’t bad at all (and thanks to Adam Lindsay for the title catch).

Now sitting in front of a new Core i7 MacBook Pro, one of the first compilations I wanted to have finished for new projects was GROMACS 4.5.5. As my procedure for compiling GROMACS 3.3.3 had been a highly-traveled page, I wanted to provide a brief summary of my successful 4.5.5 compilation.

A Few Piece Of Info

1. XCode

This used to be disc-download and install, now it’s available as a free download from the App Store (1.57 GB download, so plan to do something else while you wait for the download).

2. Homebrew

Having Homebrew installed in Mountain Lion made the installation of FFTW easy and OpenMPI trivial once gfortran was equally trivially installed. Therefore, to make your life easier, I can’t recommend a Homebrew installation enough. For additional install tweaks, I followed the following page: gist.github.com/1860902

Installation Procedure

1. Download gromacs 4.5.5

…and place it in your home folder (will go to Downloads most likely, drag it to your home folder for ease of building).

2. Extract into your home holder

…with a double-click, making ~/gromacs-4.5.5.

3. brew install fftw

With the install of Homebrew, you’ll simply run the following from a terminal window and produce the following output:

brew install fftw

==> Downloading http://www.fftw.org/fftw-3.3.2.tar.gz
######################################################################## 100.0%
==> ./configure --enable-single --enable-sse --enable-shared --disable-debug 
--prefix=/usr/local/Cellar/fftw/3.3.2 --enable-threads --disable-fortran
==> make install
==> make clean
==> ./configure --enable-sse2 --enable-shared --disable-debug 
--prefix=/usr/local/Cellar/fftw/3.3.2 --enable-threads --disable-fortran
==> make install
==> make clean
==> ./configure --enable-long-double --enable-shared --disable-debug 
--prefix=/usr/local/Cellar/fftw/3.3.2 --enable-threads --disable-fortran
==> make install
/usr/local/Cellar/fftw/3.3.2: 34 files, 13M, built in 2.7 minutes

4. brew install gfortran

If you don’t install gfortran FIRST and try to install OpenMPI, you’ll get the following error in Homebrew:

==> Downloading http://www.open-mpi.org/software/ompi/v1.6/downloads/openmpi-1.6.tar.bz2
######################################################################## 100.0%
Error: This formula requires a fortran compiler, but we could not find one by
looking at the FC environment variable or searching your PATH for `gfortran`.
Please take one of the following actions:

  - Decide to use the build of gfortran 4.2.x provided by Homebrew using
        `brew install gfortran`

  - Choose another Fortran compiler by setting the FC environment variable:
        export FC=/path/to/some/fortran/compiler
    Using an alternative compiler may produce more efficient code, but we will
    not be able to provide support for build errors.

So don’t. Installing gfortran will produce the following:

brew install gfortran

==> Downloading http://r.research.att.com/tools/gcc-42-5666.3-darwin11.pkg
######################################################################## 100.0%
==> Installing gfortran 4.2.4 for XCode 4.2 (build 5666) or higher
==> Caveats
Brews that require a Fortran compiler should not use:
  depends_on 'gfortran'

The preferred method of declaring Fortran support is to use:
  def install
    ...
    ENV.fortran
    ...
  end

==> Summary
/usr/local/Cellar/gfortran/4.2.4-5666.3: 86 files, 72M, built in 39 seconds

5. brew install openmpi

This is what allows you to use all cores on your machine and is not in the default XCode install.

brew install openmpi

==> Downloading http://www.open-mpi.org/software/ompi/v1.6/downloads/openmpi-1.6.tar.bz2
Already downloaded: /Library/Caches/Homebrew/open-mpi-1.6.tar.bz2
==> Using Homebrew-provided fortran compiler.
This may be changed by setting the FC environment variable.

==> ./configure --prefix=/usr/local/Cellar/open-mpi/1.6 --enable-ipv6
==> make all
==> make install
/usr/local/Cellar/open-mpi/1.6: 674 files, 21M, built in 5.9 minutes

6. cd gromacs-4.5.5

7. ./configure –enable-float –enable-mpi

You’ll produce output such as found in: 2012august29_gromacs455_configure.txt

You’ll also get two odd errors at the end of the ./configure run that do not affect the rest of the procedure:

./configure --enable-float --enable-mpi

...
./configure: line 29242: sort: No such file or directory
./configure: line 29239: sed: No such file or directory

So ignore them.

NOTE: If you’ve been going by my 3.3.3 procedure and used…

./configure --enable-mpi --enable-double

You’ll get the following error when you try to run make:

Making all in include
Making all in .
make[2]: Nothing to be done for `all-am'.
Making all in types
make[2]: Nothing to be done for `all'.

...

/bin/sh ../../libtool --tag=CC   --mode=compile mpicc -DHAVE_CONFIG_H -I. -I../../src -I/usr/include/libxml2 -I../../include -DGMXLIBDIR=\"/usr/local/gromacs/share/top\"   -O3 -fomit-frame-pointer -finline-functions -Wall -Wno-unused -msse2 -funroll-all-loops -std=gnu99 -MT genborn_sse2_double.lo -MD -MP -MF .deps/genborn_sse2_double.Tpo -c -o genborn_sse2_double.lo genborn_sse2_double.c
 mpicc -DHAVE_CONFIG_H -I. -I../../src -I/usr/include/libxml2 -I../../include -DGMXLIBDIR=\"/usr/local/gromacs/share/top\" -O3 -fomit-frame-pointer -finline-functions -Wall -Wno-unused -msse2 -funroll-all-loops -std=gnu99 -MT genborn_sse2_double.lo -MD -MP -MF .deps/genborn_sse2_double.Tpo -c genborn_sse2_double.c  -fno-common -DPIC -o .libs/genborn_sse2_double.o
genborn_sse2_double.c:931: internal compiler error: Segmentation fault: 11
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
make[3]: *** [genborn_sse2_double.lo] Error 1
make[2]: *** [all-recursive] Error 1
make[1]: *** [all] Error 2
make: *** [all-recursive] Error 1

So don’t do that, either. The proper flag is the enable-float.

8. make

This will produce the output available for download at: 2012august29_gromacs455_make.txt

9. make install

This will produce the output available for download at: 2012august29_gromacs455_make_install.txt

10. make links

This will produce the short piece of output reproduced below.

cd /usr/local/gromacs/bin && programs=`ls` && cd /usr/local/bin && \
	for i in $programs; do \
	   (test ! -f $i && ln -s /usr/local/gromacs/bin/$i . ; exit 0); \
	done

And with that, you should be able to run all programs from a terminal window.