To run, MStrat needs some information given in few input files such as:
The data file contains the description of accessions for each variable and allows several individuals per accession. The separator is the blank. So, avoid any space within alphanumeric data.
The variable file gives the heading of the columns of the data file. Different types of variables are defined according to their use in the MSTRAT program. Variables that are used in the optimization of the richness are called active variables. Target variables are not used in the sampling but are used to calculate the diversity score of the core collection. The variable file could be easily modified with the Tcl interface
The kernel file gives a subset of compulsary accessions in the core collection. The sampling procedure optimizes the diversity criterion of sub sample individuals that complete at best the diversity of the kernel set. Kernel core accessions can be, for example, historical or reference accessions. The list of kernel core accessions can be built up with the Tcl interface.