# author:  sjn
# date:    Mon Sep 17 14:47:06 PDT 2007
# project: compression
# scripts: jarch (john stam archive) and gchr (grep chromosome)

Your input file should already be sorted per sort-bed or all bets are off.

jarch accepts 3 or 5 column BED.

make-jarch wraps jarch and allows 4-column WIG to be converted to jarch format.
Of course, gchr still extract 5 column BED in this case.  make-jarch also takes
the standard BED file as input too.

The ID column is excluded in this archiving scheme.  Upon extraction, gchr will
create id-1, id-2, etc. values for the 4th column (iff the input file had 5
columns.  In the case of 3-column BED, you will receive 3-column BED output).
The counting (id-1, id-2, etc) starts over for each chromosome.

One can retain the 4th (or any other) column by cutting it out separately,
zipping it up and adding it to the final .jarch file (which is really just a
standard unix tar archive file).  When make-jarch is used, it will retain the
4th column for your automatically if you specify --keepID.
