Technology-driven biomedical research at CeMM Research Center for Molecular Medicine & MedUni Vienna #cancer #immunology #bioinformatics #AI #singlecell #CRISPR
Vienna, Austria
- Tweets 381
- Following 1,370
- Followers 2,901
- Likes 218
We are recruiting PhD students in ML/AI methods & biomedical applications via the @ELLISforEurope PhD Program (ellis.eu/news/ellis-phd-prog…). Keywords: Perturbation modeling, generative AI, synthetic biology, cancer immunotherapy, precision medicine. Details: medical-epigenomics.org/file…
3
4
529
Debugging @Nature: Nature’s Reporting Summary PDF has a technical problem: You can’t disable the “Plants” section. Lots of papers have this empty section – not only in Nature, but also in other NPG journals. Turns out this is a bug in the journal's PDF template that can be fixed.
1
8
1,070
HIRING: We recruit PhD students via the AITHYRA-CeMM PhD Program. Seeking 3 types of candidates: (1) experimental systems biology in immunity or cancer; (2) bioinformatics & biomedical ML/AI; (3) tech-dev & bioengineering (CRISPR, single-cell, spatial, cell programming etc) (1/5)
2
6
14
1,672
🧬CellWhisperer introduces a chat-based way to explore scRNA-seq data. By enabling natural language analysis, it bridges biologists and bioinformaticians—paving the way for AI-driven bioinformatics assistants. (10/11)
1
7
1,463
📚 We trained on >1 million bulk & pseudo-bulk transcriptomes with textual annotations that we AI-curated from GEO & @CELLxGENE Census. Our training data is open source and useful for developing multimodal biomedical AI models and future bioinformatics research assistants. (8/11)
1
8
1,420
🪄How does CellWhisperer work behind the scenes? We trained a multimodal AI that links transcriptomes and text, enabling free-text search and annotation of RNA profiles. And we connected this model to an LLM that we fine-tuned into a chat assistant for transcriptome data (7/11)
1
7
1,384
🚀We also validated CellWhisperer’s chat-based analysis with conventional bioinformatics. CellWhisperer was >4x faster (and 10x cooler 😊). Our recommendation: Use CellWhisperer for dataset exploration – but statistics is still important to ensure rigor & reproducibility (6/11)
1
9
1,583
🆕 The CellWhisperer paper (doi.org/10.1038/s41587-025-0…) includes several new analyses beyond our 2024 bioRxiv preprint (biorxiv.org/content/10.1101/…). For example, we used CellWhisperer for an AI-guided analysis of human organ development (5/11)
1
1
12
1,812
🔬 You can easily query large transcriptome datasets for your favorite biological process using CellWhisperer. Just open Tabula Sapiens (cellwhisperer.cemm.at/tabula…) or GEO (cellwhisperer.cemm.at/geo/) in CellWhisperer & type your query into the chat box – for example “infection” (4/11)
1
1
14
1,980
🔍 We investigate one of the identified cell clusters by selecting the cells & prompting CellWhisperer with ‘Describe these cells in detail’. This interactive workflow is enabled by seamless integration of the CellWhisperer AI chat box into a version of @CELLxGENE Explorer (3/11)
2
12
2,040
⚙️ To get started, let’s find cells by typing into the CellWhisperer chat box. For example ‘Show me structural cells with immune functions’. CellWhisperer scores each transcriptome by how well it matches this textual query and colors by query match (red: high, blue: low) (2/11)
1
2
14
3,873
🗨️ Just published in @NatureBiotech: Our CellWhisperer AI enables chat-based analysis of single-cell sequencing data. You can talk to your cells & figure out the biology without writing any computer code. Paper link and annotated walkthrough in the thread below (1/11)
12
184
683
64,920
⚕️ Our CELLFIE platform supports clinical translation of CRISPR-boosted CAR T cells. For example, to avoid the DNA double-strand breaks introduced by CRISPR knockout, we performed a tiling base-editing screen across RHOG and identified promising gRNA for clinical testing. (11/13)
1
83
🔥 What’s next? Our discovery of strong combined effects for RHOG & FAS knockout underlines the potential of synergistic gene edits for boosting CAR T cell function. We thus integrated combinatorial screening into CELLFIE, using the Blainey lab’s CROPseq-multi method. (10/13)
1
84
🔬 From a technical perspective, we are excited how our new in vivo CROP-seq method improves gRNA detection (reading from an mRNA transcript as in nature.com/articles/nmeth.41…) and reduces experimental noise (by using UMIs), which enables larger screens with fewer mice. (9/13)
1
1
111
💪 We also observed prolonged survival for FAS knockout CAR T cells, likely because these cells are less effective at killing each other (“fratricide”). Combining RHOG & FAS knockout, we obtained more & better CAR T cells, which further improved survival in leukemic mice. (8/13)
1
1
3
2,470
🔍 RHOG is a small GTPase involved in cell signaling. How does it influence CAR T cells ? We found that RHOG knockout increases the proliferative capacity of CAR T cells and helps them retain a highly functional state with reduced exhaustion and enhanced memory phenotype. (7/13)
1
99
🐁 We performed extensive in vivo validations and found that RHOG knockout CAR T cells achieve strong reductions in cancer cell numbers and prolonged survival in an aggressive mouse model of human leukemia, with consistent results across different CARs and T cell donors. (6/13)
1
88
🐭 But not everything that makes CAR T cells proliferate or kill better in vitro translates into more effective therapies. For scalable validation in mice, we conducted pooled in vivo CRISPR screening and observed strong positive effects of RHOG, PRDM1, and FAS knockouts. (5/13)
1
1
90
🩸 Using CELLFIE, we conducted 58 genome-wide CRISPR screens, with readouts for CAR T cell proliferation, target cell recognition, activation, apoptosis & fratricide, and exhaustion. The screens identified known genes (PD-1, CTLA4, TIM3, TIGIT etc.) and promising new hits. (3/13)
1
132
⚙️ We developed CELLFIE (“cell engineering for immunotherapy enhancement”), a CRISPR platform to make & test gene-edited CAR T cells at scale. CELLFIE supports in vitro & in vivo screens with various clinically relevant readouts, plus combinatorial & base-editing screens. (2/13)
1
1
127
🧬 CAR T cells demonstrate the power of engineered cells as therapeutics. But they fail for most patients. Can we make them better by gene editing? Our paper in @Nature presents a CRISPR platform for optimizing immunotherapies & discovering boosters of CAR T cell function. (1/13)
1
11
20
2,174
🤝Teamwork by: Peter Traxler*, Stephan Reichl*, Lukas Folkman, Lisa Shaw, Victoria Fife, Amelie Nemc, Djurdja Pasajlic, Anna Kusienicka, Daniele Barreca, Nikolaus Fortelny, André Rendeiro, Florian Halbritter, Wolfgang Weninger, Thomas Decker, Matthias Farlik, Christoph Bock (8/9)
1
2
706
🗺️In summary, this study provides a blueprint of epigenetic & transcriptional dynamics and regulator functions underlying macrophage immune responses. We found it particularly useful to combine and integrate multi-omics time-series with high-content CRISPR screening. (7/9)
1
1
378
💡Biological detail: EP300-mediated repression of interferon-stimulated genes (ISGs), validated genetically & pharmacologically. Proposed mechanism: The histone acetyltransferase EP300 counteracts HDAC activity required for BRD4 availability for transcription elongation. (6/9)
1
1
420
🤖We used machine learning to infer functional similarity maps of transcriptional regulators from the CROP-seq data, establishing a broadly applicable method to dissect transcriptional programs. (5/9)
1
1
401
✂️To disentangle causes and consequences, we performed high-content CRISPR screening (CROP-seq + CITE-seq) to perturb 135 regulators during Listeria infection. Our hits include: PU.1, JAK-STAT proteins, splicing factors (SFPQ, SF3B1) & epigenetic regulators (EP300, SMC1A). (4/9)
1
1
423
🧬Integrative analysis revealed strong differences between interferon-driven (IFN-β/γ) and pathogen-driven (Listeria/LPS) trajectories. Many immune genes showed "epigenetic potential": pre-established open chromatin ready for rapid expression upon stimulation. #Epigenetics (3/9)
1
2
472
🦠Pathogens & cytokines trigger macrophage receptors and induce immune gene expression. We challenged mouse macrophages (BMDMs) with 6 immune stimuli (Listeria, LCMV, Candida, LPS, IFN-β, IFN-γ) and profiled genes (RNA-seq) and chromatin (ATAC-seq) over six time points. (2/9)
1
1
2
819
🛡️How do macrophages tailor their defenses to different pathogens? Our new paper in @CellSystems combines dense multi-omics time series with high‐content CRISPR screens (CROP-seq) to map the regulatory landscape underlying macrophage immune responses. #Immunity #Screening (1/9)
2
38
150
15,093
🔍 While vision language models (VLMs) for digital pathology converse about H&E images as a whole, SpotWhisperer provides fine-grained, spot-level resolution. Consequently, SpotWhisperer outperformed SOTA VLMs (PLIP, CONCH) in the prediction of tumor regions and cell types. (3/6)
1
2
189
⚙️ Our SpotWhisperer method predicts spatial transcriptomes from standard H&E images with DeepSpot, and it uses CellWhisperer transcriptome-text embeddings for natural-language conversations about the cells and their transcriptomes. (2/6)
1
1
3
236
🔬 Toward histopathology 2.0: spatial transcriptomes inferred from routine diagnostic H&E images + a chat interface for cell-resolution histopathology through English language. (1/6)
1
11
36
2,975
AI in medicine ready for takeoff! 🤖🚁 We flew a drone through the Institute of Artificial Intelligence @MedUni_Wien. Join us for the trip and watch our team at "work". Credits: @muronglizi, @QuantifiedRob, @R_Bednarsky & the entire crew. #AI #biomedicine #innovation #makingof
2
8
948
CeMM's new Research Report is online - a piece of art as always (link below). Each group selected a mathematical formula to symbolize their research and visualized it in a live group picture. We picked Shannon entropy, as biomedicine is all about information (and sometimes messy)
ALT Christoph Bock Lab - Group Picture - CeMM Research Report 2024
1
1
8
564
Examples of our recent research:
1. Immunology: Immune regulation by non-immune cells (nature.com/articles/s41586-0…)
2. Cancer: Epigenetic tumor heterogeneity (nature.com/articles/s41591-0…)
(2/4)
1
1
1
279
HIRING: We offer several PhD student positions in the CeMM PhD Program. Looking for 3 types of candidates: experimental systems biology in cancer & immunity; bioinformatics & biomedical ML/AI; tech-dev & bioengineering. Interdisciplinary & collaborative lab. Details below (1/4)
1
4
21
4,293
We are recruiting a Tenure-track Assistant Professor "Machine Learning in the Life Sciences" at the Medical University of Vienna. We will nominate the top candidate for a € 1.8 million starting grant by a Viennese foundation. Deadline: Dec 31. Details: medical-epigenomics.org/file…
11
19
3,226
🚀 Outlook: CellWhisperer shows how natural language makes data exploration more fluid & creative. Let's build the future of biomedical data analysis with “AI assistants you can talk to”! How about: Interactions by voice, virtual reality integration, agent-based automation? (9/9)
1
15
2,746
📝 Internally, CellWhisperer uses Geneformer for transcriptomes and BioBERT for text, and it integrates their embeddings into a shared space using the CLIP approach. And we finetuned Mistral 7B to chat based on the embeddings, combined with the model’s biological knowledge (6/9)
1
4
2,906
📚 The hardest part was to assemble the training data. We derived 1,082,413 pairs of bulk and pseudo-bulk transcriptomes with coherent textual annotations from GEO and @CELLxGENE Census – a heroic effort of our GPUs, delivering 1000s of hours of LLM-assisted data curation (5/9)
1
1
7
3,029
🪄How does CellWhisperer work behind the scenes? We trained a multimodal AI that links transcriptomes and text, enabling full-text search and annotation of scRNA-seq profiles. And we connected this model to a chat AI that we fine-tuned to talk about the transcriptome data (4/9)
1
5
3,150
🔍 We investigate one of the identified cell clusters, selecting its cells and prompting CellWhisperer with ‘Describe these cells in detail’. This interactive workflow is enabled by seamless integration of the CellWhisperer AI chat box into a version of @CELLxGENE Explorer (3/9)
1
1
4
3,511
⚙️ To get started, search for cells by typing into the CellWhisperer chat box. For example ‘Show me structural cells with immune functions’. CellWhisperer scores each single-cell transcriptome by how well it matches this query text and plots the cells (red: high, blue: low) (2/9)
1
5
4,506
🗨️ WANNA TALK TO YOUR CELLS? Try out CellWhisperer – our new multimodal AI that turns single-cell RNA-seq analysis into a conversation. No coding needed, just chat in plain English. Short walkthrough below. Web app & bioRxiv preprint linked in the thread. Let's dive in! (1/9)
12
237
880
117,680
A week since our 2024 lab retreat, and work on some fresh projects is already in full swing. Thanks everyone for contributing and @muronglizi & @R_Bednarsky for the video (no drones were harmed...)
1
1
14
1,925
We are recruiting a PhD student in ML/AI methods and biomedical applications, within @ELLISforEurope & @humancellatlas. Keywords: Perturbation modeling, generative AI, synthetic biology, cancer immunotherapy, precision medicine. Great environment in Vienna medical-epigenomics.org/file…
1
7
24
4,066
Ready for an expedition into the unknown? We are recruiting two postdocs through the CeMM Postdoc Program. Our group combines ML/AI in bioinformatics, CRISPR & single-cell technology development, and biomedical applications in cancer & immunity. Details: cemm.at/fileadmin/user_uploa…
ALT This is our group picture from the CeMM Research Report 2022. Can you spot Charles Darwin? More background here: https://cemm.at/news/n/cemm-research-report-2022
9
26
3,540
Here’s the direct link to access the Research Briefing without a journal subscription: rdcu.be/dGnH0. And the link to the full paper (open access): nature.com/articles/s41590-0…. And a brief paper walkthrough on Twitter: nitter.net/BockLab/status/1783065…. With @MFarlik & @Nik_Fortelny
WHAT KEEPS IMMUNE CELLS AWAKE & ATTENTIVE? It’s their neighbors – stimulation from the tissue environment. We profiled epigenomes & transcriptomes of 12 JAK-STAT mutant mice & found surprisingly strong effects at homeostasis. Paper just out in @NatImmunol: nature.com/articles/s41590-0…
1
2
594
Thanks @NatImmunol for highlighting our paper with a Research Briefing, explaining the broader relevance with a bit of backstory and quotes from a reviewer and the editor. The paper is open access, but the Research Briefing is not – please use the direct link in the comment below
1
3
11
1,424