Chronological Lab Notebook

• Cool mold on Di's rose water dye project
• Gene gun shots with eyGFPuv vector: duckweed, lots of def +ve. Algae, fair few glowy bits. Water fern, a few +ves it looks like (but lots of other fluorescence so harder to tell). Onion epidermis, no +ves but was near-empty gun. Carrot callus, bit of a mess, contam, some yellowy glowy bits but nothing as obvious as I'd like. Want to try again on fresh carrot.
• Tomato (purple) indoors: first flower looks like starting to become a fruit
• Sweet peas outside: eating a few every day
• Blueberries: first fruit yesterday
• Tried protein extract on rosella and artichoke same as the passion fruit et al, no miraculin-like stuff, no water sweetness effect.
• Some creeping woodsorrel seedlines developing, identical development of plant bodies so far, cool to see.

Someone online claims artichoke makes water taste sweet, and blame a protein although they also say the effect survives cooking.
I bought an artichoke but even chewing on the raw stem for a while I don't really notice the effect.
I chopped up some stem and extracted with 70% ethanol and gentle heat, with a pinch of citric acid.
Running a TLC place with ethyl acetate gave a pink smudge for the chlorophyll, re-running with 70:20:8:2 EA: Ethanol : water : vinegar gave some blue-fluorescing smudges for (presumably) various phenolics, the front-runner of which also quenches UVC and is a likely candidate for chlorogenic acid, which the lit claims is present in large quantities in the stems of some cultivars and blames for the sweetening effect (specifically, it taste sweet when it is washed away). Tasting various concentrated versions of the extract, both plain and neutralized, didn't really get me much sweetness I could notice 🤷 I'll try cooked artichoke tomorrow, although my general opinion of this plant matches my feelings on this experiment: too much work for ~no gain :)

Tried a tazer-based electroporation test a while ago but mostly messed it up. Voltage (with 1-2cm gap) in about the right range but oscillating for 50-100ms.

Some white poppies in with the orange ones (california poppies). Apparently these have been studied and it's usually a specific mutation (76 bp deletion in PSY1A). Collected material from both a white plant and an adjacent orange, also tagged the white to later collect seed pod if possible. Some leaf matter dried and labelled, some in Zymo DNA shield (~200mg plant material in ~800uL liquid)

Primers so I don't forget:

PSY-Gap-F TCAAGCAACGACGGAGAGTA
PSY-Gap-R CCTTGCCTGCGAATATGTCT

PSY-Flower-F AAATCAACTCACAAATATTCTTAGAGATGTC
PSY-Flower-R GCCCTGCCTGTGCTAGC

Tries various gene gun shots
- Onion: no conclusive results
- Tobacco. Nothing obvious, although maybe a few yellowish cells in amongst the bluer fluorescence
- Duckweed: some definite transformations IMO!
- Noticed a stray algae tfm in the duckweed batch, repeat on a bunch of algae gave +ve results!!

Shots were weaker than I remembered; turns out the CO2 cylinder was mostly empty. Re-doing some with fresh cylinder in the gun.

Digging around in the new ESM release, I found some very close matches to miraculin in some edible plants, notably passion fruit, Roselle, Butterfly pea flower and a few trifolium species (red and zigzag clover). Ordered the first three online in various forms to test, and while waiting I tried a protein extraction with white clover from the yard:
- 5g of leaves crushed in 20ml water with 1.5g salt for ~0.6M. FIltered
- 80ml cold ethanol added, sit overnight in fridge
- Precipitate spun down, re-suspended in ethanol, spun down again, left to dry for a bit then suspended in water

I did a blind taste test with a ml on the tongue left for a bit then spat out - no change in citric acid taste between one vial prepared as above and another treated the same but then left for a while in boiling water to denature the proteins. I also tried grapefruit juice and tobasco with no noticeable difference. Maaaybe the citric acid was less harsh and lingering after the extract.

I know white clover wasn't even in the list but a few other trifolium species were. Conclusion: this is a workeable protein extraction workflow to try with the other candidates, but if there was miraculin-like protein here it wasn't in enough concentrations to affect flavour that I could detect.

Lot more colorful now. Also, sampled some sea water on the weekend, DNA ordered for miraculin project, lots of other stuff paused for now

No sign of RUBY in tomato cotoleydon bits, agro growth around 3/19 (so PPM dip somewhat effective at killing that compeltely) but also lots of dead tissue, white/brown margins on all but some bits of green left, no callus yet. Also no RUBY in the leaf injection sites I tried, maybe agro was not happiest (was an old plate that prompted this test)
pBILIv2 in the cloning strain on a plate with some iron supplementation looking a little yellow but still not lots of color. Published a video on the whole thing. Where I dropped some extra iron-roch liquid there's more color, the liquid was brownish but maybe not that strong? Certainly color seems stronger there than when I first applied the drops. So maybe it is input-limited. But not convinced since the only iron I had was a weird supplement pill from D

50g flower heads yielded 20g petals
+ 300ml water, simmer or sit in hot water for 30 mins, strain
+ 150g sugar + 1.5g agar + 1g citric acid + 0.5g Vitamin C

Next time I'd up the agar or use pectin, this was syrupy rather than set. Could also use more tartness, maybe a tablespoon of lemon.

PBILIv2 tfm into BL21 worked but now developing satellite colonies, tried a re-stream and also a streak onto a plate with iron supplementation.

Biliverdin plasmid v2 in the cloning strain it came in, vs plain e. coli. Some color - but not much, and this is after a while. I transformed it into BL21 yesterday, successfully by the look of it (plenty of nice small colonies this am) - will wait to see if they make more color, and also test iron supplementation.

This tweet wondered about a 'decaf pill' to add to drinks. I though about proteins or something that would bind (but you'd need 25g or something), enzymes to break it down (but you'd need to work in hot + acidic conditions and be food safe). But e.g. activated charcoal binds caffeiene, would that work? Test:

• Two identical ~80ml portions of instant decaf, with 100mg pure caffeine added (10ml @ 10mg/ml).
• Coffee filter with activated charcoal bed prepared, briefly washed (not enough, it turns out)
• One coffee passed through this twice then filtered through a clean filter to get rid of at least some of the charcoal dust
• re-heated to same temp, small cups used for blind taste test
• 7uL spot on TLC plate of each, run with ethyl acetate, compared to known standards with 254nm illumination

Results: TLC shows a significant reduction in caffeine, but also less polyphenols etc (judging by glow under 356nm UV torch). I found the non-filtered reference unpleasant and bitter, I actually guessed it was the filtered version and something had gone wrong! The filtered 'decaf' version tasted smooth but bland, I preferred it although this is an unrealistic caffeine concentration for coffee and there was no milk etc. (edit: added milk, closer to a tie (both bad)).

I bet you could end up with some kind of tea bag form factor with better washed activated charcoal, that you add to coffee or whatever for a while before drinking. Maybe good in a pinch if the only options are caffeinated and you don't mind a bit of flavour change...

Edit: fitting a quick curve it looks like only 50% reduction, less than I guessed by eyeballing (although likely about as accurate tbh)

I typically consume ~20-40mg caffeine per day in tea I'd guess. Today I'm trying a '5-hour Energy' drink with 230mg of caffeine. TLC confirms it does have tons of the stuff! So far:
- BP/HR spiked to 136/72 and 86bpm soon after drinking it
- Reaction time is worse (308ms peak, 290ms now vs a baseline below 270)
- Typing speed is worse, mostly due to extra errors (say, 15wpm penalty and I don't type particularly fast to start with)
- Beat Saber was interesting - I could maybe react faster, but also recovered worse from errors. Negative overall efffect imo (although it's hard to judge since any introspection distracts!)
- I'm not noticing any strong symptoms besides maybe a slight jitteriness.
(Experiment ongoing when I log this, TODO update)

Was disposing of a tobacco plant (had been gene gun testing on it). Lightly mashed a few leaf chunks in a little ethanol, to get a bright green extract. Ran a TLC plate with ethyl acetate and a big spot (10+ uL) of the extract. Nothing visible under UV. A few grains of lye and a sprinkle of KMnO4 in water as a stain brought out lovely colors though :) A second run with some ethanol + water added to the EA and a smaller dot gave a streak rather than a nice spot, the plain EA was better.

Not guaranteed the main spot is nicotine but given that it's, what, 1-2% by dry weight of the leaves, soluble in ethanol, seems pretty likely I would guess. r.f. 38mm/46mm = 0.83 ish.

Not recommending replication, if you do, be careful - this is a highly soluble toxin that migrates across skin easily. I was goggled and gloved and less cavalier than usual, and successfully maintained my "have never experienced nicotine" lifetime streak. Leaves, remaining extract etc. were carefully disposed of. I'm documenting this experiment partly since I realised after the fact that scientific curiosity might not be the first motive one might assume if they found evidence of extracting and testing alkaloids haha.

Cool iridescence with light as the right angle in these cells 🌈 diffraction effect I assume. Edit: see https://journals.asm.org/doi/10.1128/aem.07339-11 figure 4 for good comparisons

I've isolated a nice orange yeast that was growing on a neglected old plate.
I received my third plasmid design, another attempt at green. I streaked some out from the glycerol stock they provided yesterday, but this morning noticed only a few small colonies at the start of the streak... and then realized: I had forgotten to change the vector from the default (pUC57 with ampicillin resistance) - not the kan-resistant vector I wanted! Ah well. I have some old AMP plates in the fridge from last year, which I've streaked it out on, and if those are too old I'll make up some more with my last few uL of AMP solution. And then think about subcloning into a kan vector to match the other colors perhaps? Or just make art on plain LB and rely on them all not losing color too fast.

Shot some carrot callus in TC and a strip of carrot. Only one red spot which scraped off, something besides a transformed cell I suspect. And the callus got pink yeast growing on it. (I left it on air purifier overnight to dry because of family stuff, assume contam from that)

More pics of duckweed with RUBY, transformed with agrobacterium. I still need to try a fresh attempt, planning to split them in hopes of transforming more meristems, but even the rough and ready way I did these at least yields some partial transforms and a couple of fully transformed bits! Hopefully over successive generations we can weed out the mosaic mixes and get pure, but if I can up the initial transformation efficiency that'll help. How these were done:
- Agro + tfm mix in a syringe
- Add some dwe fronds
- Pull back and release plunger a few times
- Rinse a few times
- Into nutrient mix to recover + grow
- Wait.

Inspecting the leaf I shot with DE + CaCl2 + RUBY DNA has a few tiny red spots, hopeful???

Well, I tried growing out the e.c. with my biliverdin plasmid. No obvious color in the colonies (maybe a little more yellowish than normal?) but the agar over time has shifted color. I think as I feared the biliverdin is diffusing out, and it's more bruise-brown than obvious green in the LB agar. Might need to try operation treefrog.
Also, this is growing out the cloning strain I got from genscript. The plasmid DNA tubes they sent looked empty, with the dry ice I suspect the DNA was freeze-dried in the (no parafilm) tubes. I added sterile water, and tried out my spec method to see if DNA was present, seems like probably? I did a transform but got one (1) colony so something funky going on. Next trying liquid culture of the stuff I already have, and leaving things for longer to see what happens.