Jesse Engreitz

Dissecting the impact of transcription factor dose on cell reprogramming heterogeneity using scTF-seq - Nature Genetics This study introduces single-cell transcription factor (TF) sequencing, a single-cell barcoded and doxycycline-inducible TF overexpression approach that reveals dose-sensitive functional classes of TFs and cellular heterogeneity by mapping TF dose-dependent transcriptomic changes during the reprogramming of mouse embryonic multipotent stromal cells.

🧵1/ Excited to share our new paper introducing a new #singlecell assay: scTF-seq, a high-throughput single-cell approach to explore how transcription factor (TF) dose shapes cell identity and reprogramming outcomes. 🔗 www.nature.com/articles/s41... Big congrats to the entire team @EPFL & @SIAT_China

Poising and connectivity of emergent human developmental enhancers in the transition from naive to primed pluripotency In primed human pluripotent stem cells (hPSCs) resembling post-implantation epiblast, numerous lineage-specific enhancers assume the poised chromatin state, co-marked by H3K4me1 and Polycomb-associate...

Many enhancers that drive tissue-specific gene expression are already connected to gene promoters in human pluripotent cells.

In a new preprint, we share some clues about when, how, and why this happens!

www.biorxiv.org/content/10.1...

HHMI adopts Plan U journals.plos.org/plosbiology/...

Really excited to share our latest work led by @mattiaubertini.bsky.social and @nesslfy.bsky.social: we report that cohesin loop extrusion creates rare but long-lived encounters between genomic sequences which underlie efficient enhancer-promoter communication.
www.biorxiv.org/content/10.1...
A🧵👇

Excited to share our first preprint! We developed an image-based pooled screen to uncover regulators of HP1 condensates and discovered a link with intronic RNA and RNA processing. 👏 Congrats to all authors, especially Matthew, Shaopu & Chris!

We are excited to share GPN-Star, a cost-effective, biologically grounded genomic language modeling framework that achieves state-of-the-art performance across a wide range of variant effect prediction tasks relevant to human genetics.
www.biorxiv.org/content/10.1...
(1/n)

Functional maps of a genomic locus reveal confinement of an enhancer by its target gene Genes are often activated by enhancers located at large genomic distances, and the importance of this positioning is poorly understood. By relocating promoter-reporter constructs into thousands of alt...

✨Exciting news: the main story of my PhD is out in Science!

Together with Christine Moene @cmoene.bsky.social, we explored what happens when you scramble the genome—revealing how Sox2’s position shapes enhancer activation.

📖 Read the full story here: www.science.org/doi/10.1126/...

Kinetic organization of the genome revealed by ultraresolution multiscale live imaging Genome function requires regulated genome motion. However, tools to directly observe this motion in vivo have been limited in coverage and resolution. Here we introduce an approach to tile mammalian c...

Congrats to my friends in the Boettiger lab for this really beautiful live imaging work. A big leap forward in understanding the dynamic side of genome organization. www.science.org/doi/10.1126/...

Functional maps of a genomic locus reveal confinement of an enhancer by its target gene Genes are often activated by enhancers located at large genomic distances, and the importance of this positioning is poorly understood. By relocating promoter-reporter constructs into thousands of alt...

Functional maps of a genomic locus reveal confinement of an enhancer by its target gene by the @basvansteensellab.bsky.social

www.science.org/doi/10.1126/...

And a big thank you to support from:
@novo-nordisk.bsky.social
@igvfconsortium.bsky.social
NIH-NHLBI
@americanheart.bsky.social

18/18

Thanks to Lars Steinmetz + @argschwind.bsky.social for developing the original TAP-seq method and collaborating on this study

Thanks to Gene Katsevich and Tim Barry for developing SCEPTRE and collaborating to explore how best to apply it to enhancer perturbation data

17/

This was a huge team effort —

Judhajeet + Evvie + Dulguun led the development DC-TAP-seq and design + execution of the random screens

James + Evvie led analysis of random screens

Maya + Andreas compared the effects to models and teased out indirect effects

Congratulations all!

16/

Enhancer-targeting CRISPR screens at coronary artery disease loci suggest shared mechanisms of disease risk To systematically identify causal genetic mechanisms that confer risk for coronary artery disease (CAD) in GWAS loci, we mapped genome-wide variant-to-enhancer-to-gene (V2E2G) links in vascular smooth...

e.g. see preprint from the Quertermous lab using DC-TAP-seq to target elements containing variants for coronary artery disease in smooth muscle cells

The high statistical power of these experiments will be very important for finding effects in GWAS loci

www.medrxiv.org/content/10.1...

15/

We are excited to help you set these types of experiments in new systems and expand these data 10- to 100-fold in the next few years to better understand regulatory elements, improve predictive models, and interpret genetic variants.

14/

We hope that these tools are useful for you! This study presents our most complete toolkit to date for designing, conducting, and analyzing regulatory element CRISPR perturbation studies.

Code, protocols, data — all available now!

13/

Overall, these observations were consistent across the 2 cell types (K562 and hiPSCs), suggesting that they are likely to be more general beyond the favorite workhorse cancer cell line.

12/

These unbiased CRISPRi datasets will help to evaluate predictive models (stay tuned for results for scE2G and ENCODE-rE2G)

Here, we show that this evaluation must account for the magnitude of effect sizes, frequency of indirect effects, chromatin states, and gene class.

11/

Housekeeping genes appear to have similar frequencies of distal enhancers as non-housekeeping genes, but the effect sizes of these enhancers is ~2-fold weaker.

This is consistent with previous results suggesting that the promoters of housekeeping genes are less responsive to distal enhancers

10/

17% of regulatory elements corresponding to sites that bind CTCF only (no/very low H3K27ac).

The large frequency of these sites (likely, CTCF binding sites that may regulate 3D contacts) has been missed in some previous studies due to selecting elements with high H3K27ac.

9/

Nearly half of significant effects were likely to be indirect
– including nearly all of the examples of ‘up-regulation’.

So, CRISPRi is not, for example, finding lots of silencing elements.

8/

Most effect sizes were in the range of 5-10% — much smaller than effect sizes observed in previous studies.

This was not due to technical differences but rather differences in statistical power and element/gene selection bias.

7/

We found 145 significant element-gene pairs (out of 4,711 tested with good statistical power for 15% effect sizes)

The properties of these interactions differed from previous studies in a few important ways:

6/

To address this:

• We CRISPRi ~1,000 randomly selected elements in 25 loci in each of 2 cell types

• We developed DC-TAP-seq to improve guide capture and get high capture for genes of interest

• We developed a statistical power framework to ensure power for 15-25% effects on gene expression

5/

Limitations of existing datasets:

1. They often selected ‘interesting’ elements (e.g., high H3K27ac) or genes (e.g., transcription factors)

2. They have largely focused on 1 cell type (K562 cells) 

3. Statistical power was limited due to cost constraints 

4/

But, existing datasets have key selection biases that could skew our view of regulatory elements:

3/

1. Collect 1000s of CRISPR perturbations 2. Develop predictive models 3. Apply models to link risk variants to genes 4. Identify exceptions and iterate

Background: We and others have previously used CRISPRi to perturb thousands of candidate regulatory elements and measure their effects on expression.  

These studies have yielded insights about regulatory element function and enabled us to build predictive models like ABC and ENCODE-rE2G

2/

bioRxiv - An unbiased survey of distal element-gene regulatory interactions with direct-capture targeted Perturb-seq

New preprint from our lab!

What can we learn about the properties of gene regulatory elements by CRISPR’ing a random set of accessible sites in human cells?

Find out here: www.biorxiv.org/content/10.1...

👇

1/

Genetic and epigenetic screens in primary human T cells link candidate causal autoimmune variants to T cell networks - Nature Genetics Massively parallel reporter assay in primary human CD4+ T cells and bulk and single-cell CRISPR-interference screens identify candidate causal variants linked to autoimmune disease risk that modulate ...

Excited to finally present the lab's latest work in defining candidate causal genetic variants that drive autoimmune diseases and their effects on primary human T cell expression and function! www.nature.com/articles/s41...

E2G E2G is a tool based on the Open Targets Platform for predicting enhancer-gene interactions.

@riyavsinha.bsky.social @jengreitz.bsky.social sky.social @anshulkundaje.bsky.social y.social have made the ENCODE-rE2G data available to browse through the E2G portal, a custom-built extension of the Platform 👇

We plan to further integrate their data 👀

e2g.stanford.edu

Thanks to support from @igvfconsortium.bsky.social NNF and the Applebaum Foundation

Looking forward to your feedback!
👉 github.com/kundajelab/e...

8/8