Pinned
Jess Smith
Jess Smith
@nanojess.io
I do bioinformatics stuff with @fulcrumgenomics.com. I do garden stuff at home in Somerville, MA.
146 followers428 following97 posts
Jess Smith reposted
Belay Diagnostics needed a pipeline they could release and audit as Summit™ moved toward clinical testing. Fulcrum helped put automated tests, version control, and separate development, staging, and production environments in place. Read the case study: resources.fulcrumgenomics.com/hubfs/Fulcru...
> That’s because people tend to study—and protect—fluffy, charismatic, or large plants and animals. “We love those things, but they’re not necessarily the most important” Don't tell my dog! He's pretty sure he's important baynature.org/2026/09/30/s...
Interesting biology and also the kind of study where you want to DOUBLE TRIPLE SUPER CHECK your methods because this is the sort of thing that could easily be an artifact. I think they did a good job in this paper, but similar studies beware!
Herpesvirus mRNAs consistently carry longer poly(A) tails than host and other virus transcripts, suggesting a widespread strategy that may enhance virus gene expression 🧬🦠 #Herpesvirus #RNA #MolecularVirology 📄 https://doi.org/10.1371/journal.ppat.1014341 👤 EVBC member: Daniel Depledge
Just going to put this here
My research over the last several years has convinced me that organized tenants can transform our broken politics. But some folks don't yet discern the possibilities of tenant power. This piece offers insights on the real possibilities of power from the margins. www.marketplace.org/story/2026/0...
This is intriguing. I also liked this quote: > There's a radical amount of divergence in the experiences, needs, and tooling of "software engineers". Sadly, it's easy to forget this, and end up in silly arguments. sketch.dev/blog/lightwe...
Every time I read a paper like this I just think about what a PITA it's going to be to design off-target NGS assays that are suitable for regulatory review.
Donor-complementary prime editing enables precise kilobase and library-compatible DNA insertions - @fudanuniversity.bsky.social www.nature.com/articles/s41...
> with AlphaFold3 performing best overall and for the third complementarity-determining region of the heavy chain I mean, that's the ball game, and we basically already knew this but it's always good to check! academic.oup.com/bioinformati...
The headline is a little over-the-top, but this part is really worth considering: They typically cite no papers published between the mid-'90s and early 2020s at all...Why does this happen? Because GenAI models cannot get behind publishers’ firewalls... www.chronicle.com/article/ai-i...
We sure have been talking about these working for a while.... www.statnews.com/2026/08/19/m...
Not my normal beat, but my friend wrote this interesting article about intra-party gerrymandering. Important to consider for those of us who live in deep-blue or deep-red cities/states. papers.ssrn.com/sol3/papers....
> RNA-targeting CRISPR systems are commonly evaluated by RT-qPCR, but guide RNA binding can confound these measurements. We show that crRNA alone produces apparent knockdown without reducing target RNA abundance, whereas RNA sequencing remains unbiased...
Guide RNA binding induces reverse transcription bias during RT-qPCR analysis of RNA-targeting CRISPR/Cas systems https://www.biorxiv.org/content/10.64898/2026.08.16.745121v1
> But machine-learning your way to happiness with those numbers is very likely impossible, due to the huge numbers of variables in the assay conditions and indeed, in biological systems themselves. www.science.org/content/blog...
Jess Smith reposted
This paper is over a year old but it's so cool. I love when people use DNA-mediated technology to get a more complete biological picture. www.nature.com/articles/s41...
Jess Smith reposted
DivRef is a resource for including common human variants and haplotypes in CRISPR off-target searches. I rebuilt the generation workflow to make its inputs, assumptions and population choices easier to inspect and change. blog.fulcrumgenomics.com/p/a-crispr-o...
I've been using Typer more lately and I'm liking it. That's it! That's the post. Not every post can be 🔥🔥🔥🔥 typer.tiangolo.com
We cannot let China bet us in button-having. I want buttons back!!
Jess Smith reposted
This is cool! I love techniques that can give us insight beyond identifying bases. Nucleic acids deserve better than their reduced representation
What? What does this even mean! I have mad respect for plant people - so much more complicated than mammals
Avocados come in 2 varieties - A-types start each day female and switch to male midday. B-types do the reverse. In work just published from the last chapter of my PhD, we show this system evolved 40+million yrs ago and is regulated by alleles of 1 transcription factor www.pnas.org/doi/10.1073/...
Good news! It's important to sometimes read the good news and be happy
NEW | California just reached a world-first solar milestone Over HALF (51%) of electricity generation across the whole month of May came from solar power. Battery storage is the big difference maker, enabling California to push the share of solar generation to this level.
Okay this is so cool. File under: yet another thing I pessimistically thought would never work
Jess Smith reposted
I interviewed 99-year old Joan Birman who must be the oldest author on a major breakthrough paper. As an added bonus: her team beat the guys trying to use AI! Please read and share about this human triumph: www.scientificamerican.com/article/how-... #science #math 👩🔬
I like this take because it implies I get to keep my job! > AI is an extraordinarily powerful accelerant whose value scales directly with the expertise brought to bear on it. It is a springboard, not a replacement. www.nature.com/articles/s41...
Jess Smith reposted
That magical time of year when the cones burst into full bloom www.universalhub.com/2026/magical... #Boston #Allston #CommAve
We definitely need to be able to measure LLM performance, but that will inevitably cause some "teaching to the test." I'm not sure how big of a problem this is though? Clearly the technology continues to improve, at least in my hands. dylancastillo.co/posts/pelica...
Damn, this is how I learn the Earth's core is waaaayyyyyy hotter than I thought. What's going on in there?
Earth’s core may be as much as 1000°C cooler than previously thought, according to a new technique for measuring the temperature of molten iron under extreme pressures. Learn more: https://scim.ag/4yIOLkf
Good work, guys*! www.broadinstitute.org/news/low-cos... * guys in the gender-neutral way