CellProfiler Beta (2.2.0) has been released!

CellProfiler Beta (2.2.0) has been released!
http://cellprofiler.org/releases/


We’re releasing early and often

If you’ve been following us for the last several years, you may have noticed that we’ve released very infrequently
This is going to be changing in a big way. We will now release CellProfiler every six weeks. You’ll have three choices:

Nightly- this is the cutting edge build that has passed our tests. If you asked us to fix a bug and we did, this is probably what you want.
Beta - every six weeks, we’ll take a build and make it our beta version. If you want the cutting edge and want to help us by testing the latest, this is what you want.
Stable - it’s been used by the early adopters and their experience gives us confidence and you should have some confidence that this version will be an improvement for you.

Other news

  • We’ve hired the group’s first true software engineer, Allen Goodman(@allen_goodman) . He’s bringing even better software practices to the group. He comes to us with some very impressive skills and experience: he was a senior engineer at Chef working on distributed computing and brings a wealth of knowledge about operating systems, deployment and packaging.

  • The cellprofiler-announce email list is retiring. Going forward, we will send announcements like this to all registrants of our online Forum. If you’d like to continue receiving updates, look to the Forum for all CP-related release announcements.

  • CellProfiler is finally moving to 64-bits on the Mac. This means that you can use larger images. Your analyses will also run faster.

  • BSD License - We’re changing the license for CellProfiler. If you’re an end-user, CellProfiler is still free. If you’re a developer, this means there are fewer restrictions or requirements if you change or add to CellProfiler and you can use CellProfiler as part of a commercial product.

  • Workspace viewer - Hate having those windows popping up in Test Mode? Close them all and turn on the workspace viewer. You can composite images from any module, objects and measurements as you step along in Test Mode and see a total picture of what’s going on. The workspace viewer even lets you zoom into a region, change something in your pipeline and run again and the picture stays zoomed!

  • CellProfiler users are changing the world! CellProfiler has been used in recent translational research to identify therapeutic targets in Ebola, compound screens for drugs to treat rare genetic diseases, Tuberculosis and Leukemia (see http://cellprofiler.org/impact.shtml) Got your own good story? Let us know what you are up to!

5 Likes

Cool! Where can I find the changelog for the most important improvements?

Hello!
I am trying the 64-bit Windows version of the 2.2.0 release on Windows 7 but I could not install it because it claims that I don’t have java installed. However, I just installed java8 and Oracle is happily telling me that i successfully did so. However after installing also the java jdk and setting the JAVA_HOME environment variable to the jdk folder CellProfiler did install and could be started. Is it necessary what I did?

Hi Christian,

This sounds exactly right. I queried @LeeKamentsky and @allen_goodman to make sure, and we are planning to clarify the requirements (Issue here). Thanks!

You need to install the 64-bit version of the JRE. It’s actually surprisingly difficult to do and I have struggled to find permanent links. If you “install Java” from a 32-bit web browser, it will install the 32-bit version of the JRE. Amazingly, most every web browser is still 32-bit.

The following is a URL to the downloads page as of today: http://www.oracle.com/technetwork/java/javase/downloads/jre8-downloads-2133155.html . You should pick the Windows x64 JRE from the list. I am not so sure this is a stable URL.

1 Like

Good question! Older release notes are here, and when they are compiled for the latest release, we’ll post those there too. It’s been 1.7 years(!!) since a true release, it’s safe to say that there are LOTS of improvements.

Dear Alisonkozol,

Thanks a lot for so fancy tool.

But I have a little problem on installation of CP for Mac X86. It reports error as the picture shows

Did I do anything wrong, please?
I have the former version installed. Mac is X86 for sure.

Thanks a lot in advance!

@cs61tyce Thanks for the report!

  • What version of OSX are you running?
  • Did you drag the dmg into your Applications folder first, or did you run it directly from the installer dmg? If the former, can you try installing into /Applications and then running?

@David_Logan Thanks a lot for your replay.

Now it works.
I updated OS to EI Capitan, and downloaded JVM for X64, and then, it works!

Hi,
Great that the “bleeding edge area” is active again. Alas, there don’t seem to be any Linux builds. I’m still running an old 2.1.2 build that I downloaded in August (cellprofiler-2.1.2-468.20150813.433.f03b31fSNAPSHOT.x86_64.rpm). Will newer Linux builds start appearing soon again?
Thanks,

  • Damir

Hello,
I was able to run CP 2.2.0 on my Mac (OS X, version 10.10.5) after installing 64-bit version of the JRE.
Unfortunately I encounter the following problem, while using RunImageJ macro for bandpass filtration:run(“Bandpass Filter…”, “filter_large=40 filter_small=5 suppress=None tolerance=5 autoscale”);
All other ImageJ macros I am using in the pipeline are running without any problem.

Here is the error details:

Traceback (most recent call last):
File “/Applications/CellProfiler.app/Contents/Resources/lib/python2.7/cellprofiler/gui/pipelinecontroller.py”, line 2848, in do_step
File “/Applications/CellProfiler.app/Contents/Resources/lib/python2.7/cellprofiler/pipeline.py”, line 2095, in run_module
File “/Applications/CellProfiler.app/Contents/Resources/lib/python2.7/cellprofiler/modules/run_imagej.py”, line 700, in run
File “/Applications/CellProfiler.app/Contents/Resources/lib/python2.7/cellprofiler/modules/run_imagej.py”, line 805, in do_imagej
File “/Applications/CellProfiler.app/Contents/Resources/lib/python2.7/imagej/macros.py”, line 112, in run_batch_macro
File “/Applications/CellProfiler.app/Contents/Resources/lib/python2.7/javabridge/wrappers.py”, line 62, in
File “/Applications/CellProfiler.app/Contents/Resources/lib/python2.7/javabridge/wrappers.py”, line 139, in __call
File “/Applications/CellProfiler.app/Contents/Resources/lib/python2.7/javabridge/jutil.py”, line 855, in call
File “/Applications/CellProfiler.app/Contents/Resources/lib/python2.7/javabridge/jutil.py”, line 822, in fn
JavaException: Macro canceled

Would you please help me solve the issue?
The same pipeline is running without any problems on Windows 7.
Thank you.

Comments @LeeKamentsky? I haven’t tested RunImageJ on 2.2.0 yet.

If your installation runs some ImageJ macros but not others and you are
seeing a Java exception when running one of them, it’s likely that the
diagnostic info is in the Java exception which should be logged as well. I
think that should be in the Mac logs, if not, you can start CellProfiler on
the Mac from a terminal window and you’ll see the log output in the
terminal window. If you have CellProfiler installed normally, this will be
/Applications/CellProfiler/Contents/MacOS/CellProfiler.

Could you send us the pipeline for completeness? I will try running the
bandpass filter on an image via a CellProfiler pipeline and see what I get.

1 Like

I ran the attached pipeline successfully. I think I need your pipeline to continue further.
bandpass.cppipe (4.2 KB)

Hello,
Thank you for the fast response. The pipeline is attached bellow.
I was testing it several times and it seams the error occurs over every 3 or 4 sets of images on this particular imagej macro.
Screen_Total_Test_G3.cpproj (2.0 MB)

Looking at your pipeline, I think you need one last RunImageJ module at the end that closes any open images. You can try the File > Close All command or the corresponding run(“Close All”); macro - I think those should work, but didn’t test. Each time you input the currently active image into ImageJ, it adds another image to ImageJ and I am guessing that after several pipeline iterations, you’re running out of memory.

If you could test that for me and report back, it would be great. If it fails, I’ll see if I can set up the pipeline with some random images for each of the channels, try to recreate the problem and find a solution.

I tried your suggestion to run ‘close all’ at the end of the pipeline and also to have it after each channel processing. The same error occurred after 11 set of images were processed successfully. The size of each image is ~ 14 mb.

You can also try increasing the Java Heap size (Maximum memory for Java), in File -> Preferences…

Those are pretty large. It sounds like Close All helped somewhat. There are two ways you could alleviate the memory problem. The first is to increase the amount of memory available to ImageJ. The easiest way to do this is to open the preferences dialog (under the File menu on Windows) and increase the Maximum memory for Java (it’s near the bottom of the dialog). You should restart CellProfiler after increasing this number. The second would be to add a few more RunImageJ modules with “Close All” macros at strategic places in your pipeline (perhaps even after each of your current RunImageJ macros).

I would definitely go with David’s suggestion and increase the memory to about 1024 Mb, especially if you have a new machine and then if that doesn’t work, add more RunImageJ modules.