Ben Kleinstiver

If you like transposons...
If you you love genome editing...
Or if you just like random bird animations,

we have the paper for you!

We (@kedmonds.bsky.social et al) are happy to share our work turning a songbird retrotransposon into a genome editing tool. 🐣 (1/n)

🚨 Our work on optimizing single-stranded DNA donors with Cas12a binding moieties is published at @moltherapy.bsky.social Nucleic Acids! 🧪

Since the preprint, we added two super interesting findings. Short 🧵👇 (1/5)

www.cell.com/molecular-th...

#ImmunoSky #GeneEditing #CARTcells

Less than one day to celebrate a major win for science, before returning to our regularly scheduled programming of destruction of the American scientific enterprise and capitulation..

1/10 Today in @science.org in collaboration with
the Liu group we report the development of a laboratory-evolved CRISPR-associated transposase (evoCAST) that supports therapeutically relevant levels of RNA-programmable gene insertion in human cells. drive.google.com/file/d/1I-Ub...

Genomes encode biological complexity, which is determined by combinations of DNA mutations across millions of bases

In new work @arcinstitute.org, we report the discovery and engineering of the first programmable DNA recombinases capable of megabase-scale human genome rearrangement

Baby Is Healed With World’s First Personalized Gene-Editing Treatment

A gene-editing treatment used on a 9½-month-old boy with a rare condition has the potential to help people with thousands of other uncommon genetic diseases #NBTNewsBeat www.nytimes.com/2025/05/15/h...

For more about how Rachel created this bespoke Cas9 protein, see here!

bsky.app/profile/bkle...

An incredible collaborative effort led by @kiranmusunuru.bsky.social @ahrensnicklas.bsky.social @urnov.bsky.social & others.
Congratulations to @rachelsilverstein9.bsky.social who designed the Cas9 enzyme that went into this drug.
Wishing K.J. & his family all the best 🙏

Today "a milestone in the evolution of personalized therapies for rare & ultra-rare inborn errors of metabolism"
—the 1st human to undergo custom genome editing
—from decades of NIH funded research
www.nejm.org/doi/full/10....
@nejm.org
www.nejm.org/doi/full/10....
www.nytimes.com/2025/05/15/h...

A screenshot of the termination notice showing "Outstanding Investigator Grants"

A screenshot of the termination notice with "This award is terminated effective the date of this award, due to unsafe antisemitic actions that suggest the institution lacks concern for the safety and wellbeing of Jewish students." highlighted

Yesterday, the NIH R35 “Outstanding Investigator” grant to fund scientists in my lab studying antibiotic resistance was terminated for reasons not related to the content of the science, or any actions taken by me or members of my lab

Plasmid of the Day

pCMV-T7-SpCas9(MKRCMV)-P2A-EGFP (AHK162)
Depositor: Benjamin Kleinstiver @bkleinstiver.bsky.social
Purpose: pCMV and pT7 Human expression plasmid for SpCas9 enzyme with MKRCMV amino acid substitutions

www.addgene.org/223075/

Linking CRISPR–Cas9 double-strand break profiles to gene editing precision with BreakTag - Nature Biotechnology The genome-wide landscape of Cas-induced double-strand breaks and end structures is profiled at nucleotide resolution.

The genome-wide landscape of Cas-induced double-strand breaks and end structures is profiled at nucleotide resolution go.nature.com/44DO2Cc
rdcu.be/ekKtv

Go easy on that poor old guy!!

By combining high-throughput protein engineering with machine learning, bespoke SpCas9 enzymes are identified that outperform evolution-based and engineered nucleases and base editors in human cells. www.nature.com/articles/s41... #NBTHighlight

Machine Learning Engineers Bespoke Cas9 Enzymes for Gene Editing Researchers have developed a machine learning model that predicts Cas9 proteins that can be tailored with designer properties for therapeutic use.

Thanks to @faylinphd.bsky.social from @GEN for featuring our manuscript!

www.genengnews.com/topics/genom...

Thanks to @mgbresearch.bsky.social for writing a press release about our manuscript!

www.massgeneralbrigham.org/en/about/new...

A ‘pie in the sky’ goal when we started the lab quickly became a huge half-decade collab. effort that required oodles of assay devevelopment.
Deeply grateful to our @NIHDirector DP2 award that enabled this project. 🙏
Science like this depends on visionary funding. ✊
Happy editing! 🧬

More broadly, this framework should be extensible to enrich our editor toolbox, including the use of ML for other properties of Cas9 enzymes, for other CRISPR-Cas or non-CRISPR enzymes, or to other protein domains in next-gen editors.

These are just a few of many possible applications of PAMmla predicted enzymes that can improve the safety & efficacy of genome editing.

We hope that this massive toolbox of bespoke Cas9 proteins shifts user to deploy more custom tools, especially given their advantages for genetic medicines.

Then, with Qin Liu @MassEyeAndEar, we utilized PAMmla to develop bespoke allele-selective Cas9 nucleases to treat a blindness causing mutation.
Importantly, PAMmla reversed the biology of Cas9, now able distinguish against a canonical NGGG PAM & instead effectively target an NGTG PAM.

To enable users to predict their own bespoke Cas9 enzymes, Rachel developed an in silico directed evolution pipeline and worked with @lucapinello.bsky.social's lab to build a website that permits prediction and navigation of novel Cas9 proteins *in real time* - check it out!
pammla.streamlit.app

Working with Dr. Suk See De Ravin @NIH, we demonstrated that PAMmla enzymes are highly effective as base editors to correct mutations in patient-derived primary cells while also minimizing off-targets.
No need to use PAM-relaxed enzymes like SpG anymore - use more specific PAMmla proteins instead!

Because PAMmla-predicted enzymes are more selective for their PAMs, they encounter fewer off-target sites around the genome.
This results in safer enzymes that minimize unwanted off-target editing.

Testing PAMmla enzymes in human cells led to encouraging results - as we'd hoped, PAM altered enzymes enabled higher levels of on-target editing for nucleases and base editors, while minimizing editing at unwanted PAMs in ways previously not possible with PAM-relaxed enzymes.

We validated that PAMmla could accurately predict the PAM profiles of Cas9 enzymes, both previously unseen before in nature or in our training data sets, and those of previously engineered SpCas9 PAM variant enzymes (developed by us or others).

@rachelsilverstein9.bsky.social then utilized machine learning to investigate all of these proteins including the rarer enzymes, training a PAM machine learning algorithm (PAMmla) that could then predict the PAM requirements of all 64 million Cas9 proteins varied at these 6 amino acids.

The selections yielded two major phenotypes: enzymes that can target 'NGG' sequences (akin to wild-type SpCas9), or those that can tolerate a relaxed 'NGN' PAM (similar to previously developed generalist enzymes). But hidden in this dataset were some more rare Cas9s enzymes..

We performed saturation mutagenesis of 6 amino acid residues in Cas9, subjected this library of enzymes to a large selection experiment (against different PAMs), and then utilized scalable assays to sequence & characterize ~800 Cas9 proteins (thx to HT-PAMDA; @russelltwalton.bsky.social )

We asked a simple question: Can we optimize experimental assays to identify and deeply characterize a sufficiently large quantity of Cas9 proteins, permitting us to then utilize #machinelearning to predict the properties of the remaining millions of enzymes not seen in our training data?