This work was led by 𝐍𝐨𝐥𝐚𝐧, with contributions from 𝐓𝐚𝐭𝐢𝐚𝐧𝐚. Thanks also to many group members for testing and providing feedback on the method. Funding from Kreftforeningen, NFR, Familien Blix Fond, and iCAN 🙏

📖 In the manuscript, we walk the user through a tutorial on how to run the SiSaNA pipeline, interpret the results, and obtain high-quality figures summarized in a single HTML file as output. 🌐

This work was led by 𝐍𝐨𝐥𝐚𝐧, with contributions from 𝐓𝐚𝐭𝐢𝐚𝐧𝐚. Thanks also to many group members for testing and providing feedback on the method. Funding from Kreftforeningen, NFR, Familien Blix Fond, and iCAN 🙏

📖 In the manuscript, we walk the user through a tutorial on how to run the SiSaNA pipeline, interpret the results, and obtain high-quality figures summarized in a single HTML file as output. 🌐

Read their side-by-side Q&A on how international mobility builds both skills and community:

🔗 projects.au.dk/nordic-embl-...

@mkuijjer.bsky.social @afarkkila.bsky.social @embl.org @mimsumea.bsky.social @dandrite.bsky.social

This work presents a major milestone by the group, led by Tatiana Belova, with contributions from current and former group members. Funding sources incl the Norwegian Research Council, Norwegian Cancer Society, iCAN, and NCMBM 🙏 Code is available in Snakemake 🐍 pipeline: github.com/kuijjerlab/P...

Finally, we mapped TFs modulating PD-L1, with their regulatory profiles linked to survival. Recurrent TFs included both known modulators (incl with/without known binding sites) and TFs with yet unknown role in PD-L1 regulation, highlighting candidate targets for modulating immune checkpoint activity

Strikingly, PD1 signaling heterogeneity associated with survival in about half of these (12/23) cancer types, but only associated with immune infiltrate in a subset (5/12, 👇), suggesting that, in many cancers, dysregulation may arise from tumor-intrinsic regulatory programs

This motivated us to map the regulatory heterogeneity per cancer type with our tool PORCUPINE, analyzing the complete networks. This identified several recurrently heterogeneously regulated pathways (categories and pathways from top category 👇), with PD1 signaling dysregulated in 23/33 cancer types!

We reconstructed sample-specific networks with our NetZoo tools PANDA and LIONESS: nearly 10k networks for 33 cancer types. Visualizing and clustering these networks on Gene Targeting Scores indicated that, for most cancer types, networks stratify patients distinctly from expression data (e.g. 👇)

Poster discussion ongoing, with our iCANDOC PhD student Panos presenting benchmarking of single sample network subtyping 📜