Dan says:
Regarding the blued text especially.
Huh. I suppose this is roughly a verification of what I believed and poked in here as text, unfortunately.
It's remarkable how ( hopefully its unintentional ) misleading the WWW stitched around databases is for these alleged magic rabbits-from-hats systems composed of WWW programs, fat PDF's and CSV ( or more fashionably, TSV ) files.
I worry I am rattling on about the obvious, but you might notice I am interested in numbers that create milestones in time, or other ways to estimate actual, as opposed to synthetic, progress.
One way I thought about it lately is the H.S.-tuna-fish-to-ATP benchmark. The baseline is, if I gobble back a tuna fish sandwich and a map is developed from plate to ( ATP and 10 A.A); and we assume the area under the curve is 95% the mass of the food; ( ex. exclude metabolics for the olives, maybe etc.). How many unique proteins are in that transformation ?
e1) < 100
e2) > 100 < 1000
e3) > 1000 < 5000
If you go super low and dare e2. think how reliable each transformation has to be characterised to win ? If many are 30 xforms long ... and 98% percent grokked. 30^ in that subset is in for failure in series is ... 0.98^30 = 0.55
So if either the deviation from the central dogma, errors of omission, mini-fraud, whatever is at 98% some are only 1/2 good. out of 30 of those, the chance ones important and completely obliterates the analysis is kind of well, certain.
So beyond P.R. to get funded, get queen-kissing-ops with Nobel Prizes, etc Mother nature has made this not medium hard but really, really hard.
I wonder if this kind of thinking has utility, or the general approach is: "Shut up and keep shoveling", that is the hard way is the only way... period.
A.I. programs kind of allow some more deviation to allow the unknown, but expecting a lot of them seems to me more misrepresentation, maybe.
I'd suggest a lot more humility in goals, and specifically NAMING of projects etc would help slightly. To explicitly discuss hoping to get from 2% to 10%, not a vague sci-fi ready story that's some mix of wishful thinking and fraud.
Regs,
Daniel B. Kolis
my ref: BioBricks, blog, 28 Dec 2022, JonC