Interesting thing (based on my own self thinking which may absolutely not apply beyond ) about the use of AI in manuscript or proposal reviews. As the writer of said manuscript or proposal, I don't need three paragraphs on what we said. I need a paragraph on why you think a specific request is key.

Just enough clouds for a quick splash of color

If you make a mutation leading to a metabolite building to higher levels creating changes in growth, etc. Does this inherently mean autotoxicity? Are there alternative options like allosteric flux issues, regulatory effects, etc? Or are they all autotoxicity and underneath is the mechanism?

Cleared just enough to catch the last rays.

Swirling pastels tonight

I understand putting species information in front of a gene name. But is two letters enough? For example, is AtMYC2 in Arabidopsis thaliana, Amaranthus thunbergii, Aquilegia taygetea, Aethionema thesiifolium, Alyssum taygeteum or any of the others? We probably need some other way to standardize.

So if no targeted assay finds things like imidazole or purine alkaloids in a plant species. But an untargeted LC-MS suggests they are there, what should we believe? Especially if the untargeted never comments on the fact that the targeted studies never find them?

Conundrum, if AI speeds up everything in biology, then we will need a ton more wet lab people to do the research. Unless one posits all knowledge can be obtained and tested in single celled organisms grown by a robot in an incubator. Don’t laugh you know at least one person that thinks this.

Is academics really the issue with the translation/product transition? From what I have seen it is the corporate accountants and investors that punish failure so hard that it dramatically increases the allure of the simple solution snake oil, even when it is obvious not to work.

Still bummed that a reviewer wouldn't let us use *TI and *AMP for ETI, PTI, DAMP, PAMP, HAMP, MAMP, VAMP, etc.

For todays fun and excitement it is time to work through pre-requisite appeals for a large required lecture class. Most of them are just transfer classes that campus doesn't work to automate translating to simplify effort all around.

If we as a plant community don’t conduct real phylogenies on genes, what does any ortho/paralogue translation or machine learning effort even mean?

When the best AI algorithms take days of labor and computer time to find the exact same genes a simple regression found!

Jake Gyllenhaal: I Think We're Making Progress

ALT: Jake Gyllenhaal: I Think We're Making Progress

static.klipy.com

The subtle sunset nights really let you see night and day at the same time.

Necessary and sufficient is a gold standard in genetics thought to mean key importance. Which often becomes a way to say we know it all so why do more. However, necessary and sufficient is only true in the specific background tested, so likely way more going on.

Sparkly sunset tonight, makes the pastels pop.

Fire in the sky tonight.

Interesting that the vast vast majority of mechanistic addendums to papers working on plant genomes are about metabolism rather than signaling or other processes.

New Kajala lab photo has dropped! It is such a pleasure to work with these early career researchers - each bring their talents and passion. No wonder I go to work smiling! 🇫🇮🇮🇹🇲🇽🇳🇱🇮🇳🇦🇷🇬🇹 #PlantSciences

Quite the color palette tonight.

Dream - Setting: Faculty meeting Date: Fall 2036 Topic: Reading [Prof 1] AI agents have such lovely speaking voices and can read us any text. Why should we teach reading. We can always have AI teach us if we need to learn. [Me] What the .... (wakes up) was that a nightmare or just prediction?

When it has been a month since seeing the three foot garter snake and you start wondering, hmm. Did the hawks take out all the snakes. A 4.5 foot snake skin appears. Dang things hide too well.

Baby’s first show.

Big brother, little brother.

Random thought question (or if there is data even better). Mutate 100 random TFs and mutate 100 random enzymes. Make the single gene mutations (lets not worry about epistais). Do scRNAseq or bulk transcriptomics. Which set of mutations will alter the accumulation of more transcripts?

From our FY26 joint CMM call: Dawn Chiniquy is using high-throughput bacterial genetics & targeted genome engineering to study how plant-associated bacteria capture and tolerate rare earth elements, aiding recovery strategies. @berkeleylab.lbl.gov Full portfolio: https://jointgeno.me/CMMFY26

Oddly as someone originally trained as a biochemical geneticist in the classical fashion, I think I am getting inured to gene based studies and the idea that just because genes and connections have been found that we learned something. Or is that the original training coming through?

Found this art work in Salt Lake City. Funky art deco columbine rendering. But cannot figure out the artist. Anyone in the botanical/art intersection have an idea?

We are deeply saddened to share that Michael Freeling passed away on 7/20/2026 after a year-long battle with cancer. Our deepest condolences go to his family, friends, colleagues, and the many students he inspired. We will share additional memorial information soon.

Too paste or to sauce while editing proposals and papers?