Howard Salis

NEW PAPER led by @grace_vezeau! We rationally engineered synthetic riboswitches to carry out sense & respond for human biomarker proteins [mCRP and IL32-gamma] that indicate disease or infection. 16-fold activation without trial-and-error! nature.com/articles/s4146…

@jboysen0 Yep, we do like to collaborate with folks working in this area, eg biorxiv.org/content/10.110…

@hsalis If you wanna collaborate, they’ll chip in on some gBlocks I’m sure

Behold peasants, 10k synthetic enhancers with no wetlab verification

@ShaoBin_phy To what extent is the transformer learning evolutionary conservation? e.g. rRNA genes are highly transcribed. TFs are lowly transcribed. What happens if a microbe has a rare/unique enzyme?

Is it possible to get the transcriptome of any sequenced microbe without doing the experiments? Happy to introduce TXpredict, a transcriptome prediction tool that generalizes to novel microbial genomes: biorxiv.org/content/10.110…

Great work by Howard Salis et al. @hsalis @NatureComms nature.com/articles/s4146…

An autonomous microbial sensor enables long-term detection of TNT explosive in natural soil @NatureComms from @hsalis "sensors and memory-storing genetic circuits inside the soil bacterium Bacillus subtilis" 😲 nature.com/articles/s4146…

@alexcampreis @GallowayLabMIT @AkosNyerges RNA-binding proteins are plentiful in eukaryotes. The same might be true, but most "unstructured" RNA likely bound by something. RBP-mediated protection from RNA decay could be regulated by signals, environment, etc.

@GallowayLabMIT @AkosNyerges @hsalis The Salis lab work was in bacteria. I wonder if it's been studied as deeply in eukaroytes?

Somewhere I recall reading that longer UTRs generate less stable transcripts, whereas shorter UTRs are more stable. Anyone got a reference on that?

@hhlee @SynBio1 I was joking with my UGs that, to be more like tech, biotech needs to make 90% of its $$ from ad revenue. Taking your daily statin? Watch an ad. Going to the ER? Watch 10 ads. Undergoing a heart transplant? Watch a thousand ads.

@SynBio1 Hurts so good. How bout an ads-based model. The more clicks you get, the cheaper it is to access. Editors and Lab both get royalties. True face revealed.

Biotech VCs often complain that scientist-founders aren't good at product development but we should cut them some slack knowing they were trained in a culture where nature magazine thought people would pay $8.99 to rent an article on readcube

@O_Borkowski Any place is better than here. Too bad there's no "Bluesky" for the real world.

@hsalis Forget what I wrote. Bluesky is the new thing : )

Hey Synbio folks. Any interest in migrating over to Reddit? I'm tired of this place. The char limit is annoying. The trolls are bad company. And there's no back & forth conversation so what's the point? reddit.com/r/SyntheticBio…

@rosscloney See you there!

Also a chance to use this gif again

So you may have noticed this account has been quiet for a while. I've recently noticed a lot of researchers on Bluesky and I'm making the leap too: rosscloney.bsky.social

@JoshuaSchrier @SavoieLab @olexandr But p(doom)/d(EvErYtHiNg ELSE OMG WHY WTF HOW) is definitely positive.

Dinner conversation tonight with @SavoieLab & @olexandr : `p(doom)` is the wrong question. The right question is what is the sign of `d p(doom)/d (AI)` (and we agree that it is negative)

@GallowayLabMIT harness leverage apply combine integrate :)

Reminder: Just use "use". There's rarely a need for "utilize". I know folks think it is fancy 🧐🎩 but let's make the ideas fancy and the text clear and simple. I hope you find this useful. If not, feel free to not utilize it.

This is a great paper from the @hsalis lab. - Measure the decay rates of 50,000 mRNAs in bacteria. - Use biophysical models + ML to build models of mRNA stability. - Profit. And a good reminder of what's possible when one turns a biological problem into a sequencing problem!

@jrkelly @ATinyGreenCell It’s pretty easy to design around patented sequences nowadays, including proteins.

@ATinyGreenCell you think modern design tools will just let us easily design around IP anyway? that's why there's effectively no patent issues in software code -- trivial to design around most things. same for DNA code in the future?

What would unlock biotech industry to reach same scale as tech? 🧬

I've uploaded new versions of ApE 3.1.7. The Mac version didn't have the correct version number in the "Get Info" window. The Windows and Linux versions had a small bug that prevented them from running. All three errors should now be fixed.

Designing synthetic mRNAs 🧬 for optimal stability and expression? Checkout our work led by @DanielCetnar on biophysics infused machine learning approaches for delineation of mRNA degradation kinetics ⚡ out now in @NatureComms nature.com/articles/s4146…

Great work by @hsalis et al. In @NatureComms nature.com/articles/s4146…

congrats @DanielCetnar, @hsalis and coauthors on the recent mRNA stability paper in nature comms! Using MPRAs and ML to elucidate the sequence and RNA structural determinants of RNA decay is really neat stuff... paper in comments

@MarkBlenner No idea. Ambulance was on the scene.

@hsalis Did they get to vote?

So I'm inside the church where we vote & an old lady backs up her car into the brick building, slamming into the wall & causing structural damage. Then she panics and drives full tilt *forward* into a ditch, while hitting the side of a parked car. Lady is OK, no others were hurt!

Police arrive and I overhear him saying quite nonchalant "Yeah, this happens a lot ... they panic & confuse the pedals and just keep going". But she had to *change gears* idk how/why. While inside, people continue voting, and the church bake sale was thriving. #Kamala4President