Here are a set of scripts that I've been using to generate mapability
and gc-content annotation files. Drop the perl scripts in
the same directory as the bash script, then run the whole thing from
the bash script, with the arguments:

1 - chr name (matches the fasta file name (use chr5 to match chr5.fa)
2 - the average length of your reads
3 - the directory containing your per-chromosome fastas
4 - the output directory

It also requires:
- that BWA be installed and in your PATH.
- that in your fasta directory, there's a concatenated fasta file of
all the chromosomes called all_sequences.fa

Keep in mind that the initial files will be quite large, so make sure
you've got ~100GB disk space free if you're using human-size
chromosomes.