Many DNA specialized formats are delta from a "baseline" model to the actually described DNA strand. `zstd` has a `--patch-from` mode, which does essentially t

Many DNA specialized formats are delta from a "baseline" model to the actually described DNA strand. `zstd` has a `--patch-from` mode, which does essentially the same thing, computing a new file using another file (typically an older version) as a reference. When similarities are larges, it leads to a huge reduction in compressed content. I wonder what would be the performance of `zstd --patch-from=` when employing the same reference as these specialized DNA compressors.