so storing data on multiple (billions of bacteria). I assume that you would have populations of duplicates and would piece together your entire original dataset

so storing data on multiple (billions of bacteria). I assume that you would have populations of duplicates and would piece together your entire original dataset (e.g. a tar) using start and end encoding tags? How do you re-sequence multiple populations of bacteria that you just grab from a sample sitting in a -80c freezer? You'd also have to prevent them from modifying their own genome and messing up your data.