Profile avatar
sroyyors.bsky.social
Regulatory genomics, machine learning, networks, systems biology, evolution
47 posts 1,124 followers 242 following
Prolific Poster
Conversation Starter

very cool www.nature.com/articles/s41... Conservation of regulatory elements with highly diverged sequences across large evolutionary distances

How might AI transform science? DSI Director @kylecranmer.bsky.social will discuss the diverse ways AI could revolutionize scientific practice at this @uwbadgertalks.bsky.social live event in Green Bay. Join us at the Neville Public Museum, June 10, 6pm. www.nevillepublicmuseum.org/events/calen...

Our New perspective explores breakthroughs & challenges in regulatory & systems genomics: from cis-regulatory code & 3D chromatin to spatial omics & AI models. 🔗 doi.org/10.1093/bioa... #Genomics #Bioinformatics #RegulatoryGenomics

HAPPY 75th, NSF! We’re celebrating this milestone by highlighting some of NSF’s most transformative accomplishments—innovations that have shaped our world and continue to drive progress in health, technology, the environment, and beyond. Read on 🧵(1/11):

Wow, this paper on interpreting ML models by @suinlee.bsky.social and Scott Lundberg came out in 2017 and has 34k citations! Congratulations @suinlee.bsky.social and your team, what an achievement :-). arxiv.org/abs/1705.07874

Here it is! Bonsai. Now there is really no more excuse for using t-SNE/UMAP. Bonsai not only makes cool pictures of your data. It actually rigorously preserves its structure. No tunable parameters. Incredible work by @dhdegroot.bsky.social. I'm so excited about this! www.biorxiv.org/content/10.1...

In case you have not seen it 😉 Here is our latest paper! With @arnausebe.bsky.social @ianakim.bsky.social @encent.bsky.social and many more 👏🏻 www.nature.com/articles/s41...

Fantastic work Lars and team! 👏🏻👏🏻

🧪 Though a doubling of NSF's budget might seem far-fetched, that's exactly what Congress said what's needed to maintain national competitiveness in science & tech. From the bipartisan CHIPS and Science Act of 2022:

multideconv - Integrative pipeline for cell type deconvolution from bulk RNAseq using first and second generation methods https://www.biorxiv.org/content/10.1101/2025.04.29.651220v1

From a genomics & bioinformatics perspective, these new NIH policies will likely knock out the pillars of these research fields across the world. No joke. Reseachers everywhere should be aware and **non-US funders need to step up** to fill the big gaps that will appear in the coming months.

So happy to see this published now! We present TGIF, Tree-Guided Integrated Factorization, based on multi-task NMF to examine dynamics of the 3D genome organization. genome.cshlp.org/cgi/content/... with rockstar student, now Dr. Erika Da-Inn Lee.

Our new call for VIB.AI group leaders is online! Junior and senior positions available to develop innovative ML methods in biology. With professorship at CS or Medical Faculty. Deadline 14th June. DM for more info.

New blog post: let's talk about latents! sander.ai/2025/04/15/l...

Our latest work now online in Cell: Rewriting regulatory DNA to dissect and reprogram gene expression Our new method (Variant-EFFECTS) uses high-throughput prime editing + flow sorting + sequencing to precisely measure effects of noncoding variants on gene expression Thread 👇

Genomic language models: opportunities and challenges: Trends in Genetics www.cell.com/trends/genet...

Hey #genomics, #bioinformatics & #algorithms peeps 💻🧬. If you haven't seen the CfP for WABI '25 yet, check out the website wabiconf.github.io/2025/. It will be held at UMD @umdscience.bsky.social with Broňa Brejová & myself as co-chairs! Submit your exciting & late-breaking algorithmic work to WABI

Make sure to also check out the preprint on the new CREsted package: many more sequence-to-function models across different species! … and it works with HyDrop v2 data as well. Congrats to @niklaskemp.bsky.social & @seppedewinter.bsky.social & all the others working on this super cool project!

Very nice perspective on causal machine learning, perturbations + single cell omics, gene regulatory networks. nature.com/articles/s41... graph-based representations are important for getting to mechanism. This really resonates with me.

📢 new preprint alert: So so excited to share our analysis on the impact of common and rare variants on single-cell gene expression in blood, using WGS and scRNA-seq data from nearly 2,000 individuals and 5.4m cells as part of TenK10K phase 1 🧬 www.medrxiv.org/content/10.1... 🧵👇 (1/n)

ATTENTION: Do you or someone you know get fired or have a job offer rescinded because of uncertainty due to the NIH funding freeze? Please DM me! THANK YOU!

Our 11yo is telling us science dad jokes as a way to delay having to go to bed. It's working. My favorite of the evening: Why did the amoeba fail its math test? .... .... because it multiplied by dividing

This is sad not just for genomics in the US, but across the world: Eric has been an incredibly effective global champion for the power of genomics and the impact of genomic medicine.

In these unprecedented dark times for American life science and medical research (no less well beyond biomedicine), my coping mechanism is to laser focus on the unremitting and remarkable daily advances, and the fervent hope these will not be slowed or interrupted.

Vinod's team's work could have saved my wife's life, and I hope this and others like it will make a huge difference with patients with seemingly untreatable deadly diseases. This needs to be able to continue. www.mskcc.org/news/can-mrn...

Did you know that science labs work like small business entrepreneurs? Faculty hired on strength of ideas, get some startup $ to last 3-4 yrs. After that is grants- grants pay all our + our trainees’ salaries + scientific work. Funding in this country is frozen. That means scientific work stops

📄 Update on our preprint about Gene Regulatory Net (GRN) benchmarking 📄 We have included the original and decoupled version of SCENIC+, added a new metric and two more databases. Dictys and SCENIC+ outperformed others, but still performed poorly in causal mechanistic tasks. doi.org/10.1101/2024... 👇

Our new pre-print, investigating a few important questions when we train S2F models on different types of MPRA datasets. Congrats to Yilun and @xinmingtu.bsky.social www.biorxiv.org/content/10.1...

As federal funding for research is uncertain, we need to diversify funding sources for our PhD students. Here is a comprehensive list of >160 PhD fellowships and funding opportunities, mostly from private foundations and scientific societies. Download it here: research.jhu.edu/rdt/funding-...

Our latest: uncovering hierarchical design rules for cis-regulatory elements with a high-throughput screen CRE activity follows multiplicative modular rules, tuned by spacing & positioning Enables engineering synthetic CREs for predictable tissue patterning www.biorxiv.org/content/10.1...

Systematic evaluation of single-cell multimodal data integration for comprehensive human reference atlas. https://www.biorxiv.org/content/10.1101/2025.03.06.637075v1

Evaluation of deep learning approaches for high-resolution chromatin accessibility prediction from genomic sequence www.biorxiv.org/content/10.1...

A pre-trained SCVI model of our Tahoe-100M cell atlas is now available as a HubModel! It can be used for gene expression visualization, differential expression, or cell clustering. Available at Hugging Face with instructions on how to load it (see thread)

Excited to have our group represented at the CSHL Prob. Modeling in Genomics. meetings.cshl.edu/meetings.asp... Check our talk by @prakritigarg27.bsky.social and poster by @yudaliu.bsky.social on comparative regulatory genomics and gene regulatory networks!

Metascience and AI Postdoctoral Fellowships Grants of up to $250,000 (USD) over up to two years will be awarded to social sciences and humanities postdoctoral researchers who study the implication of AI for Science. Deadline: April 10, 2025 @sloanfoundation.bsky.social 🧪 sloan.org/programs/dig...