*FITTING WITH fitOSET_C We are still trying to decide how to define the “best fit” parameters and their errors, and the best algorithm for fitting. The core functionality of fitOSET_C is written in c++. *OVERVIEW* fitOSET_C fits the parameters of an OSET to a monte carlo by minimizing a chi-squared-like variable whose “bins” are the counts in each signature. The signature counts of the monte carlo are referred to as the targetvector. fitOSET_C performs the minimization from several different starting points in order to avoid local minima. Optionally, fitOSET_C can create .eps output of the best fit signatures against the target signatures. The signatures that fitOSET_C chooses to plot are not the same as the ones that it uses to do the fit, instead they are the signatures defined in MySigs in MPotato. In order to create these plots, include the .lhco.gz file for the target monte carlo as the optional argument in the call to fitOSET_C. *OUTPUT*** On output, fitOSET_C prints the best fit parameters and some error information to the screen, and stores the output in files in analysis/.

############################################ ####Quick Documentation for FitOSET_M####### ############################################ The FitOSET.pl Perl script wraps Mathematica code used for Chi-Squared minimization. It uses the arguments to determine relevant files to read from, and reads signature data, names, and variable information from multiple files, and consolidates it all for the Mathematica script to read, then calls the Mathematica script.

1. Arguments OSETStem – the common part of all OSET filenames–each process-specific filename should be of the form stem_xx.blah, and the meta-files should be meta/stem.blah.

ModelData.raw – full filename of the .raw file to which we compare the OSET. Typically in sigs/

ResultFile – file to which Mathematica output is written

2. Files read meta/stem.coefs