Source code to accompany:

A mutual information approach to automate identification of neuronal clusters in Drosophila brain images Nicolas Y. Masse, Sebastian Cachero, Aaron D. Ostrovsky, Gregory S. X. E. Jefferis Division of Neurobiology, MRC Laboratory of Molecular Biology, Cambridge, CB2 0QH, UK.

Start Matlab and paste in the following code:

tempscript=[tempname() '_fccinstall.m'];
urlwrite('https://raw.github.com/jefferis/FruCloneClustering/master/src/Easy_Install_Masse.m',tempscript);
run(tempscript);
delete(tempscript);
cd(fullfile(final_src_path,'src'));

(This saves the Easy_Install_Masse.m script to a local temporary file, runs it, removes it and then changes directory to the newly downloaded source code – all because matlab apparently cannot run a script from a url). See src/Contents.m for details of Matlab code and how to install additional dependencies of the full image processing pipeline.

If you want to keep up to date with the codebase/contribute, then you need to install git.

Then in the shell:

cd /some/sensible/dir
git clone http://github.com/jefferis/FruCloneClustering.git
git clone http://github.com/jefferis/MatlabSupport.git

In Matlab:

• Add FruCloneClustering/src and the whole of MatlabSupport to your path (IMPORTANT!)
• Run FruCloneClustering/Easy_Compile_Masse.m

See score_trace_wrapper.m for examples of querying the database of fruitless clones with a traced neuron. These examples are part of the function help that can be displayed by typing help score_trace_wrapper in Matlab.