Howard Salis @hsalis

Professor of synthetic biology & chemical engineering. De Novo DNA founder. Engineering microbes with predictive models & algorithms. salislab.net UC Riverside, CA Joined April 2009

Tweets

4K
Followers

4K
Following

260
Likes

3K

Howard Salis @hsalis

4 hours ago

@anshulkundaje @mosasaurus27 From my POV designing bacterial genetic systems with 100+ parts, mRNA / protein expression is controlled by many tunable “knobs” & we do need to understand their mechanistic seq2fcn & stochastic dynamical models including their coupling. Bacteria have long range interactions too!

1 2 2 530 0

View Details

Howard Salis @hsalis

3 days ago

@ATinyGreenCell @Moon_Synth_Bio @jrkelly So much irony here that the CCP exploits labor with 9/9/6 conditions, few protective laws, fewer clinical regs, effectively competing with capitalistic countries that protects workers & patients with a 40-hr work week, OSHA, EHS, FDA, etc. Karl Max would 🤯 looking at the CCP

1 0 2 45 1

View Details

Sebastian S. Cocioba🪄🌷 @ATinyGreenCell

7 days ago

SPLICECRAFT v1.0 IS LIVE!!!!!!!!!!!! Open your terminal and type in "pipx install splicecraft" if you want to try it out, then spam "splicecraft update" often as I push updates frequently. A labor of love for the community I adore. Enjoy! 💚

15 60 309 16K 97

View Details

Howard Salis @hsalis

2 weeks ago

@SynBio1 and 3) developing countermeasures (vaccines, treatments) so fast that any threat is mitigated before it becomes a national security issue. That removes state-sponsored attacks as existential threats. Bioterror is no longer terror-inducing.

1 0 1 36 0

View Details

Howard Salis @hsalis

2 weeks ago

@The__Taybor As a long time nanopore user, is it possible to extract the DNA that has passed through pores, using adaptive sampling to selectively pass or flick based on sequence correctness? Would be a *great way* of combining DNA assembly & sequencing in one integrated workflow...

1 0 1 176 0

View Details

Howard Salis @hsalis

2 weeks ago

@SynBio1 Specifically: salislab.net/software/build… I wrestle with this question.

1 0 1 80 0

View Details

Howard Salis @hsalis

2 weeks ago

@SynBio1 What are your thoughts on a lab open-sourcing another oligopool-based DNA assembly method? Option A: open-source it because the genie is out of that bottle already Option B: don't open-source it because improved algorithms solving this problem will add more biosecurity risk

2 0 1 225 0

View Details

Howard Salis @hsalis

2 weeks ago

I want my super intelligent AGI to be like Vic Fontaine (Deep Space 9). An emotive superintelligent hologram, but no need to lord it over the corporeals.

0 0 1 126 0

View Details

Justin Blethrow @JBlethrow

3 weeks ago

At @plasmidsaurus, we discovered our $50, 3 day TAT RNA-seq service can also detect mycoplasma contamination from several prominent species, and we've already alerted over a hundred labs to likely contamination. We're working on adding this to the service for free very soon.

Ming "Tommy" Tang @tangming2005

3 weeks ago

Your cell line sequencing data isn’t mapping well? Before blaming the aligner—check if you're sequencing bacteria instead of human. Let’s talk about mycoplasma contamination.

3 5 40 12K 14

1 9 111 18K 12

View Details

Howard Salis @hsalis

3 weeks ago

@ProfTomEllis @SynBio1 A research org that only expects to exist for <10 years will absolutely cut all sorts of corners because why not?

0 0 1 24 0

View Details

Howard Salis @hsalis

3 weeks ago

@ATinyGreenCell What's your typical LLM workflow to accelerate writing tests & harnesses?

1 0 1 139 0

View Details

Howard Salis @hsalis

3 weeks ago

@SynBio1 This is what happens when GPT plays the Reverse UNO card on your image request.

1 0 1 59 0

View Details

Howard Salis @hsalis

3 weeks ago

@SynBio1 My cat *loves* whipped cream from the can, but not if you whip it yourself w/o the sugar. It could be the texture, or perhaps he’s a mutant.

3 0 5 545 0

View Details

Howard Salis @hsalis

3 weeks ago

@rlacombe @MoAlQuraishi @gembioworkshop @iclrconf You're thinking that more states = more complexity, but it's actually not that difficult to compute. We're developing whole-cell models & solving grand canonical ensembles involving 100s of DNA binding factors to calc binding probabilities. Much easier @ isotherm + isobaric!

0 0 1 48 0

View Details

Howard Salis @hsalis

3 weeks ago

@rlacombe @MoAlQuraishi @gembioworkshop @iclrconf There are more chemical reactions taking place inside a refinery than in the human body.

1 0 0 37 0

View Details

Howard Salis @hsalis

3 weeks ago

@rlacombe @MoAlQuraishi @gembioworkshop @iclrconf Then by induction, biological systems have state functions. What's your point?

1 0 0 26 0

View Details

Convoke @Convokebio

3 weeks ago

Today we're publishing our index of unmet needs in human disease: 2443 indications scored and ranked on burden of disease, prevalence, pipeline activity, and treatment burden. We hope this will help drug developers identify overlooked medical problems convoke.bio/blog/introduci…

6 36 265 89K 283

View Details

Howard Salis @hsalis

4 weeks ago

@Avi_Marcus @ATinyGreenCell @_inc0_ Nitrogenase uses ATP and electrons to catalyze its reaction. And they are high potential electrons that require special carriers (ferredoxins).