The Demos directory contains an example from the C. elegans genome.
You have three directories to start with -- input, standard and output.

The input directory contains two files for the input of the RECON
package -- the seq_name_list_file "seqnames" and MSP_file
"elegans.msps".  The elegans.msps contains MSPs from an all-vs-all
comparison of the elegans genome which cover regions similar to Tc3 (a
transposable element in elegans), directly or by transitive
similarity.  These include both the Tc3 copies in the genome, as well
as certain Tc3-related but distinct transposable elements.

The standard directory also contains two files -- Tc3.list and
Tc3.msps.  The former file is a manually curated list of Tc3
copies/fragments in the genome.  For the more critical users, the
latter file is a manually curated list of MSPs from a search with the
canonical Tc3 against the genome, which we believe are hits to true
Tc3 copies.  You can compare this file with the "elegans.msps" to
check whether our curation is reasonable.

The empty directory of output is where you can run the RECON package
on this example (after installation, of course).  Within the output 
directory, type

../../scripts/recon.pl ../input/seqnames ../input/elegans.msps 1

The process will generate several directories to store the output.  All
these directories, except for the summary directory stores intermediate
results (see 00README).  You can find that in summary/eles, the Tc3 copies
are defined as family six.  Do

awk '$1==6' summary/eles

to get the list of members in this defined family and compare it to
standard/Tc3.list to evaluate the result of this automatic clustering.
