Stegle Lab

Meet Hendrik Mehrtens who recently joined our lab as a PhD student - shared with @mlolab.bsky.social 👋

Hendrik’s research focuses on reliable machine learning methods integrating large-scale EHR data with multimodal data like genomics data, with a focus on causality and uncertainty estimation.

Claudia Feng's Ph.D work is now out, co-led with @brittavelten.bsky.social and @steglelab.bsky.social . Large-scale CRISPRi perturbations in genetically heterogenous iPSC lines, with single cell readout - enjoy :)

@lucamarconato.bsky.social held a workshop during day 3 (@biohub.org) showing analysis and visualization strategies for large-scale 3D spatial omics datasets.

See you all at #scverse26!

@manusaraswat.bsky.social presented a poster on his work on deep learning of single-cell gene regulatory networks, applied to understand plasticity in glioblastoma.

@danaivagiaki.bsky.social gave an oral presentation on our new interpretable deep learning framework (LIVI), which enables trans-eQTL mapping at single-cell resolution. (preprint coming soon!)

Back from #scverse25 and feeling inspired by what @scverse.bsky.social has achieved over the last years. Impressive tools, a vibrant community and we even had a reunion with our two alumni Danila and Max Frank.

We also presented our latest findings from the lab 👇

The ELLIS Society welcomes its new Board. As the primary decision-making body, it will play a vital role in shaping the future of ELLIS and advancing #AI and #MachineLearning research across Europe in a time of global change.

Read the full article: ellis.eu/news/ellis-s...

Welcome to our team, Gesa Voigt! 👋

During her molecular biology studies, Gesa developed an interest in applying mathematics and ML to understand complex biological systems. Her PhD research will focus on inferring GRNs by integrating insights from population genetics and genetic perturbation data.

Please join me in welcoming Stefanie Mantz to our lab! 👋

Steffi joins us as a PhD student after completing her Master’s at Heidelberg University. She will explore gene regulation and cell differentiation using organoid models and is excited to analyze these complex datasets and develop ML methods.

MOFA-FLEX: A Factor Model Framework for Integrating Omics Data with Prior Knowledge Latent factor models are first-line analysis approaches for single- and multi-omics data, essential for data integration, alignment, and biological signal discovery. To cater for new technologies and ...

Many thanks to all authors - Arber Qoku, Martin Rohbeck, Florin Walter, Ilia Kats, Florian Buettner - and everyone who supported this work. Funding from @erc.europa.eu , @denbi.bsky.social and @dfg.de.
@dkfz.bsky.social @embl.org @oliverstegle.bsky.social

📑 Preprint: doi.org/10.1101/2025...
13/13

MOFA-FLEX PyPI Tests codecov Documentation graphical abstract MOFA-FLEX is a versatile factor analysis framework designed to streamline the construction and training of complex matrix factorisation models fo...

You’ll find all the code and documentation here:
mofaflex.readthedocs.io/stable/

pip install mofaflex

… and you’re ready to go! 12/n

Curious to build your very first MOFA-FLEX model and test it on your own data?

Check out the tutorial thread by @mlolab.bsky.social to see how it’s done: bsky.app/profile/mlol... 11/n

MOFA-FLEX is a unified framework that advances factor analysis for the omics community. Built on modular probabilistic programming, it lets researchers quickly design and deploy custom latent variable models by assembling reusable modules, fostering rapid innovation and extensibility. 10/n

3️⃣ We applied MOFA-FLEX to breast cancer data combining Xenium (166K cells; ~300-gene spatial) and Chromium (30K cells; ~3K-gene non-spatial) datasets from adjacent FFPE slides, revealing latent factors capturing global cell-type variation and transcriptome-wide spatially resolved gene programs. 9/n

2️⃣ When applying MOFA-FLEX to a multi-omic CITE-seq dataset from murine spleen and lymph nodes (N=14,870 cells), collected in two independent batches on different days, it was able to disentangle both batch and cell type variation. 8/n

1️⃣ We applied MOFA-FLEX to an scRNA-seq dataset of PBMCs from lupus patients (N=13,576 cells) treated with or without IFN-β, showing it can identify IFN-β gene programs consistent with known biology and reveal new program-linked genes (e.g., TNFSF10). 7/n

We showcase MOFA-FLEX in diverse applications: 1️⃣ robust recovery of gene programs from noisy prior knowledge in scRNA-seq, 2️⃣ disentangling technical and biological variation in multi-omic CITE-seq, and 3️⃣ spatial modelling revealing disease-linked gene programs in breast cancer. 6/n

Additionally, MOFA-FLEX adds a domain knowledge module that links latent factors to gene programs. It seamlessly integrates resources like Hallmark, Reactome, or custom sets to guide model learning and enhance the interpretability of discovered biological factors. 5/n

We teamed with @mlolab.bsky.social to develop MOFA-FLEX, addressing these challenges. Built on probabilistic programming, it unifies factor analysis extensions (flexible priors, non-negativity, supervision, alternative likelihoods), enabling declarative model design without manual engineering. 4/n

However, the rapidly evolving technological and experimental landscape demands universal, user-friendly frameworks that can be tailored to specific needs, such as incorporating spatial structure, temporal dynamics, noisy single-cell data, and domain-specific knowledge. 3/n

Advances in technology enabled parallel analysis of multiple biological layers. With the Buettner lab and others, we developed MOFA and related latent factor models to uncover key variation sources in multi-omics data. Today, factor models are first-line tools for sc and multi-omics analysis. 2/n

Are you using any of our factor models, such as MOFA? 🛵
You might’ve found it challenging to tailor them to your specific use cases - not anymore!

Introducing MOFA-FLEX: a flexible, modular factor analysis framework designed for customizable modeling across diverse multi-omics data scenarios. 1/n

Thank you to everyone who joined the Life Science Thematic Workshop at the AI in Science Summit in Copenhagen.

Led by EMBL’s Oliver Stegle and Anna Kreshuk, the workshop united researchers and AI experts to explore how advanced computational approaches are shaping the future of life sciences.

#ai

Looking forward to this collab with @moritzmall.bsky.social and Barbara Treutlein!

We’ll generate high-throughput multi-modal CRISPR data on ASD risk genes in 🧠 organoids & assembloids, and build an ML model of gene regulation dynamics to analyze the data and enable future in silico experiments.

Do you want to link genomic variants to functional gene expression - cell by cell?
SDR-seq makes it possible; an experimental method connecting coding and non-coding variants to expression in single cells.
Proud to have contributed a custom FASTQ-processing tool to this great collaborative project.

Don’t miss Manu’s poster on DeepGenoXcan, our new personalized and interpretable sequence-to-expression model, at #ASHG2025!

📅 October 16, 2:30pm-4:30pm
📍 Poster 1007T

Super excited for the 2nd edition of the #scverse conference!
A fantastic opportunity to showcase and discover innovative data frameworks for #singlecell analysis, and to meet the developers behind the scverse ecosystem. Looking forward to connecting with the community!
Register now! 👇

A scalable framework enabling the induction and spatial genotyping ➡ phenotyping of hundreds of tumors in mouse liver. #CHOCOLAT-G2P

Check out the publication for the nuts and bolts of 🍫-G2P: doi.org/10.1038/s415...

Application - Helmholtz Information & Data Science Academy

We're seeking a PhD student passionate about ML, biology, and driving impact in healthcare!

You'll be working on multi-omics factorization models for personalized medicine, co-supervised by @junyanlu.bsky.social.

👉 Project: www.helmholtz.de/assets/hidss...
✍️ Apply: hidss4health.de/application

Caption+Credit: EMBL’s AI Strategy aims to transform life sciences by integrating AI across research areas like genomics, drug discovery, and imaging. Creative team/ EMBL

🌐 EMBL’s new initiative is set to transform life science research through full AI integration – from genomics to drug discovery.

💡 Funding from the Hector Foundation is dedicated to attracting interdisciplinary talent, boosting infrastructure, training, & innovation.

www.embl.org/news/connect...