https://t.co/EHR030gO5i

Today in @NatureBiotech, we report engineered DNA recombinases with up to 53% integration efficiency and 97% genome-wide specificity at an endogenous human locus, inserting large DNA cargoes up to 12 kb for stable expression in primary human T cells, stem cells, and non-dividing

Mapping chromatin structure at base-pair resolution unveils a unified model of cis-regulatory element interactions: Cell

NYU Langone’s Rodent Genetic Engineering Lab (RGEL) is hiring a Microinjectionist! Join our team to work on cutting-edge genome engineering in mice. 🐁 Apply now:

Super cool concept for driving tissue-specific transgene expression! https://t.co/7r1TltTksm

Did it work? Hell yeah! Read our two companion preprints to learn more: https://t.co/efyQXfeAVl

This recombination fuses mouse Taf1's exons 1-24 with human TAF1's exons 25-38 to form a fully functional gene that encodes mouse TAF1 protein. The recombination also removes all marker cassettes, making the final "converted" allele scarless.

The final step of engineering is what we call "conversion". It includes introducing Cre enzyme that induces recombination between the 1st loxP site in mouse Taf1 intron 24, and a 2nd loxP site located in human TAF1 intron 24.

This feat of engineering was performed using a variant of a method we recently published called mSwAP-In (mammalian switching antibiotic resistance markers progressively for integration). https://t.co/OGh8DCdUtn

Next, we engineered mouse embryonic stem cells (mESCs), first with a loxP site in Taf1's intron 24, followed by integration of a marker cassette (a landing pad) downstream of the gene, and finally we replaced this cassette with the 78-kb human fragment.

Using CREEPY (CRISPR-mediated editing of synthetic episomes in yeast) we've substituted every amino acid that is different between humans and mice so that the expressed protein would be identical to mouse TAF1. https://t.co/e9cjbb3AKe

But we were worried that the human TAF1 protein would be incompatible when expressed in mice (TAF1 is the largest subunit of the TFIID complex, a large protein group that is essential for initiating transcription of most genes).

We assembled this region from an XDP patient-derived BAC, using the superpower of S. cerevisiae (yeast), into a delivery-ready vector.

The pathogenic SVA insertion is in TAF1's intron 32, but we wanted to generate a partially humanized mouse model, and decided to include a 78 kb human region that spans TAF1 exons 25-38, including all intervening introns and the SVA.

Our solution? A "hybrid-convertible" allele:

This is the problem we had to tackle when we engineered the first mouse model of X-linked Dystonia Parkinsonism, a disease caused by an SVA (SINE-VNTR-Alu) retrotransposon insertion in the essential TAF1 gene.

The answer is quite simple of course: make it conditional. But what if the lethality is induced by a 2.7 kb retrotransposon insertion into an X-linked gene? How would you CONDITIONALLY introduce this element?

How could one engineer a mouse carrying a LETHAL allele? A thread🧬

One step closer to an LLM that understands novel DNA🧬 𝗜𝘁𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗶𝗺𝗽𝗿𝗼𝘃𝗲𝗺𝗲𝗻𝘁 𝗼𝗳 𝗱𝗲𝗲𝗽 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗺𝗼𝗱𝗲𝗹𝘀 𝘂𝘀𝗶𝗻𝗴 𝘀𝘆𝗻𝘁𝗵𝗲𝘁𝗶𝗰 𝗿𝗲𝗴𝘂𝗹𝗮𝘁𝗼𝗿𝘆 𝗴𝗲𝗻𝗼𝗺𝗶𝗰𝘀. https://t.co/MOdUqgltN4 Congrats @andremrsantos and Matt Maurano!

Great 🧵 describing our innovative mouse model of XDP