someone already posted this idea at least 8 years ago:

> someone July 13, 2017 at 9:59 pm: if the bacteria can be made sensitive to different frequencies of light (like with rods and cones), and if the cell can be programmed to consider one wavelength a clock signal, another wavelength a data signal, perhaps it can become a cheap synthesizer for DNA fragments, optical UART to DNA bacterium

https://hackaday.com/2017/07/13/movie-encoded-in-dna-is-the-...

And then about (transcript) 1 year and 7 months later on Feb 13 2019 we have Thomas Shaddack saying 12:32 PM

>@thethoughtemporium i am nurturing a thought of light-controlled dna or rna printing. a variant on transcription, but with light pulses to energize the given nucleotide addition. it's a bit far in the left field, kind of an artificial anoparticle/"enzyme" that'd absorb at five wavelenghs, have one for each nucleotide (add to the growing chain on illumination), and one for reset (to prevent longer light intervals from making polynucleotides).

https://hackaday.io/event/163454-open-source-biology-and-bio...

Sometimes DARPA is real slow on identifying good ideas. (more than 8 years later today...) better late than never.

The most obvious route would be to first break up the task into subgoals which can be pursued in parallel:

1) achieve working ab initio and,or in silico simulation of reverse transcriptase

2) achieve working ab initio and,or in silico simulation of relevant proteins and molecules in phototransduction cascades (retinals, opsins, ...)

then:

3) analyze the phototransduction cascade in simulation and using current knowledge of known mechanisms, predict how to change wavelengths for phototransduction cascade, predict how to change end result (a conformational change, etc.)

4) analyze the reverse transcriptase and use current understanding of the mechanisms to change codon tables implemented by reverse transcriptase, analyze which conformational changes are responsible so it doesn't need the RNA input.

then:

5) test small modifications to observe shifts in wavelengths etc to verify the simulation from 2)

6) test modifications to reverse transcriptase: basically swap some elements in the usual correspondence (codon tables) between RNA and DNA bases, to verify the simulation of the reverse transcriptase

then assuming multiple teams were working either on the reverse transcriptase OR on the phototransduction:

7) form random pairs of teams and have each pair of teams try to combine their experience and workflows to achieve a single cell transducing light signals to DNA.

Oh it's much older than 8 years... here is one such proposal from 2011: https://groups.google.com/g/enzymaticsynthesis/c/6GZT8zFNOfo...

I myself was bitten by a radioactive grad student in 2008 that was obsessed with this idea at the time, and have since learned that almost every major household name lab PI has thought about this in one form or another.

> Sometimes DARPA is real slow on identifying good ideas. (more than 8 years later today...) better late than never.

Perhaps they noticed a development (breakthrough?) in an adjacent field that makes this now more interesting and/or practical/feasible.

Interesting finds, did you pull those from memory or do you have a search method?