samblaster - 0.1.26-4 main

Current "next-generation" sequencing technologies cannot tell what
exact sequence they will be reading. They take what is available. And
if some sequences are read very often, then this needs some extra
biomedical thinking. The genome could for instance be duplicated.
.
samblaster is a fast and flexible program for marking duplicates in
read-id grouped paired-end SAM files. It can also optionally output
discordant read pairs and/or split read mappings to separate SAM files,
and/or unmapped/clipped reads to a separate FASTQ file. When marking
duplicates, samblaster will require approximately 20MB of memory per
1M read pairs.