Manu Saraswat

Dived into past, present & future of human genetics with brilliant students & mentors.Grateful for the chance to present my work on personalized sequence→expression prediction and discussions with @sashagusevposts.bsky.social @bpasaniuc.bsky.social @mashaals.bsky.social @tuuliel.bsky.social & others

Evaluating the representational power of pre-trained DNA language models for regulatory genomics - Genome Biology Background The emergence of genomic language models (gLMs) offers an unsupervised approach to learning a wide diversity of cis-regulatory patterns in the non-coding genome without requiring labels of ...

Our work on "Evaluating the representational power of pre-trained DNA language models for regulatory genomics" led by @AmberZqt with help from @NiraliSomia & @stevenyuyy is finally published in Genome Biology! Check it out!

genomebiology.biomedcentral.com/articles/10....

All the best @paubadiam.bsky.social 🎉🎉

Excited to share my first contribution here at Illumina! We developed PromoterAI, a deep neural network that accurately identifies non-coding promoter variants that disrupt gene expression.🧵 (1/)

Thanks for your thoughts

I am genuinely wondering and curious what you think. It’s increasingly hard to see how academia can attract or retain top talent with offers like this, especially when industry offers 2–3x more. Something structural has to change if we’re serious about advancing AI in science.

Hi Andrea,
Thanks for sharing this, it’s an exciting and meaningful opportunity. But to be honest, the listed salary (€42–49k for postdocs) is deeply misaligned with the qualifications you're asking for- post-PhD experience, publications in top AI and biology journals, teaching, and vision.

How does tumour heterogeneity arise? How can we predict cancer cell plasticity? In 2 new studies, we trace #glioblastoma heterogeneity to a spatial cancer cell trajectory w. multimodal cell atlassing bit.ly/4mkrWgs & predict plasticity w. snRNA/ATAC+deep learning bit.ly/3FbI6Ic 🧵

Thanks a lot for the shoutout Stein. scDORI builds upon the insights from the foundational works in GRN inference from your lab - Cistopic, SCENIC, SCENIC+
Thanks a lot for your contributions. Very exciting time to be in the field 🚀

Thanks to all our collaborators, specially @lauraruedag.bsky.social who co-led the computational analysis and was the best partner in crime one could ask for. What a delight it was!
Thanks to Elisa, Tannia and Fani for leading the experimental aspects.

14/ We believe this approach can be applied to other cancers to uncover—and exploit—plasticity brakes. Get in touch if interested! #GBM #MultiOmics #CancerResearch #deeplearning #cancerneuroscience #GRNs #Cancer #singlecell

Decoding Plasticity Regulators and Transition Trajectories in Glioblastoma with Single-cell Multiomics Glioblastoma (GB) is one of the most lethal human cancers, marked by profound intratumoral heterogeneity and near-universal treatment resistance. Cellular plasticity, the capacity of cancer cells to t...

13/ Want to learn more? Read our regulatory paper www.biorxiv.org/content/10.1... and companion spatial multi-omics study www.biorxiv.org/content/10.1...
Watch out for thread from @bayraktarlab.bsky.social on mapping conserved spatiotemporal trajectories of GB

13/ Our framework unifies regulatory and spatial logic of GB heterogeneity. While our companion paper showed subclones are intermixed across conserved tissue niches, our regulatory model explains WHY—they follow the same trajectory because they're constrained by the same regulatory rules!

12/ Beyond chromatin changes, MYT1L transformed GB cells into less aggressive neuronal-like cells with: • Enhanced neurite-like morphology • Reduced tumor microtube connectivity • Decreased proliferation • In vivo: slower growth, less invasion, longer survival!

11/ Our experimental validation was striking: • MYT1L overexpression closed >80% of differential chromatin regions • MYT1L knockout reopened access to plastic fates • MYT1L directly bound and repressed regulators of other GB states• 85% of scDORI's predicted TF targets confirmed!

10/ 🔥 Can we use these GRNs to manipulate tumor identity and push GB cells into less plastic states? YES! We predicted MYT1L as the key regulatory bottleneck—a master repressor that locks cells into neuronal-like states by directly binding and suppressing the regulators of plastic states.

9/ Remarkably, our regulatory roadmap explains tumor architecture! States with easy transitions exist in close spatial proximity, while states separated by regulatory barriers are spatially distant. The regulatory rules we've uncovered directly shape how tumors are organized!

8/ This roadmap revealed striking asymmetry in GB plasticity: • OPC/NPC-like and AC-like states can easily activate multiple alternate fates • Neuronal-like states are "locked" by strong repression barriers • Transitions follow preferred directions (OPC→Neuronal easier than Neuronal→OPC)

7/ 🔑 The big question: Which tumor states can easily transition to others? We developed metrics to quantify both activation potential (what enables transitions) and repression barriers (what prevents them), creating the first regulatory roadmap of GB plasticity.

6/ The power of multi-omics: We can distinguish what a cell IS versus what it COULD BECOME. While only 16% of Topic Regulators are expressed across different tumor states, over 54% are epigenetically accessible—revealing "primed drivers" ready for activation during transitions!

5/ Applied to GB, scDORI uncovered Topics that redefine tumor heterogeneity through regulatory logic. Each Topic links specific TFs, enhancers, and target genes that work together across tumor states. We identified key "Topic Regulators" (TRs)—master TFs for each Topic.

4/ scDORI: • Scales to millions of cells • Models continuous cell state GRNs • Incorporates both activation AND repression signatures • Each cell is modeled as a mixture of regulatory Topics • Can be applied to ANY multi-omic dataset (happy to hear your feedback, separate 🧵 on soon!)

3/ To decode GB's regulatory logic, we developed scDORI—an autoencoder that decomposes multi-omic profiles into "regulatory Topics." Each Topic represents specific TF-target gene relationships, modeling cells without requiring predefined cell-types. Code: github.com/bioFAM/scDoRI

2/ Key finding: Malignant GB cells are dramatically more plastic than non-malignant cells, BUT neuronal-like tumor states show surprisingly LOW plasticity. This hints at regulatory constraints we could potentially exploit therapeutically!

1/ Glioblastoma is driven by cellular plasticity—tumor cells switching between states. With @bayraktarlab.bsky.social, we profiled >1 M nuclei with single-cell multi-ome (RNA+ATAC) from 12 GBs, capturing the full tumor heterogeneity. We asked: what regulatory mechanisms guide these transitions?

🧠 Excited to share my main PhD project! We mapped the regulatory rules governing Glioblastoma plasticity using single-cell multi-omics and deep learning. This work is part of a two-paper series with @bayraktarlab.bsky.social @oliverstegle.bsky.social and @moritzmall.bsky.social, Preprint at end🧵👇

Design principles of cell-state-specific enhancers in hematopoiesis Screen of minimalistic enhancers in blood progenitor cells demonstrates widespread dual activator-repressor function of transcription factors (TFs) and enables the model-guided design of cell-state-sp...

Out in Cell @cp-cell.bsky.social: Design principles of cell-state-specific enhancers in hematopoiesis
🧬🩸 screen of fully synthetic enhancers in blood progenitors
🤖 AI that creates new cell state specific enhancers
🔍 negative synergies between TFs lead to specificity!
www.cell.com/cell/fulltex...
🧵

We released our preprint on the CREsted package. CREsted allows for complete modeling of cell type-specific enhancer codes from scATAC-seq data. We demonstrate CREsted’s robust functionality in various species and tissues, and in vivo validate our findings: www.biorxiv.org/content/10.1...

On January 6th, 1995, my (now ex) wife and I boarded a flight from Heathrow to JFK on a one way ticket. We had two suitcases and about $900 in cash - this was everything we owned and we were moving to the US. We thought it was maybe for 2-3 years. I had visited the US once, for a conference, and