This is really neat! Are you looking at deleterious genes as well, i.e those which provide fitness defects? Seems useful to see if most genes are "background" or if some are genuinely negative.
Also - for your 11 different microbes- are those all for salt tolerance? Or are those a range of cultured bugs you expect to have at least one of the five conditions for Mars tolerance? I see radiodurans in there so I assume it's all conditions. I wonder what the expected % of E coli transformants you expect to see for any combination of conditions... I bet astronomically low. Super cool work. Excited to see what you guys do next!
We're starting to look at fitness defects as well, but the dynamic range is much lower. When something drops out of the library, it can be hard to tell if it's due to noise or negative fitness! You need really big bottlenecks and very deep sequencing depth, or else you'll just lose the less fit stuff to the noise of low sampling counts.
We figured these 11 microbes are a good panel for us to explore this technique with. There's some good diversity and variable capabilities in there, and we do plan to apply it across some more conditions soon! For polyextreme conditions we're thinking that combining multiple fragments iteratively might be the path to go, instead of hoping for the perfect fragment that can simultaneously address multiple issues.
This is really neat! Are you looking at deleterious genes as well, i.e those which provide fitness defects? Seems useful to see if most genes are "background" or if some are genuinely negative.
Also - for your 11 different microbes- are those all for salt tolerance? Or are those a range of cultured bugs you expect to have at least one of the five conditions for Mars tolerance? I see radiodurans in there so I assume it's all conditions. I wonder what the expected % of E coli transformants you expect to see for any combination of conditions... I bet astronomically low. Super cool work. Excited to see what you guys do next!
We're starting to look at fitness defects as well, but the dynamic range is much lower. When something drops out of the library, it can be hard to tell if it's due to noise or negative fitness! You need really big bottlenecks and very deep sequencing depth, or else you'll just lose the less fit stuff to the noise of low sampling counts.
We figured these 11 microbes are a good panel for us to explore this technique with. There's some good diversity and variable capabilities in there, and we do plan to apply it across some more conditions soon! For polyextreme conditions we're thinking that combining multiple fragments iteratively might be the path to go, instead of hoping for the perfect fragment that can simultaneously address multiple issues.