Manlio De Domenico

Networks are #complex and their dynamics often look chaotic. But we can reconstruct latent spaces where their behavior becomes strikingly regular, revealing functional organization across biology, society and technology.

How? 👉 rdcu.be/eSsqn 🧪🧠🦠🧬🌐

Kudos to Andrea, @dzanc.bsky.social & Sebastiano!

Networks are #complex and their dynamics often look chaotic. But we can reconstruct latent spaces where their behavior becomes strikingly regular, revealing functional organization across biology, society and technology.

How? 👉 rdcu.be/eSsqn 🧪🧠🦠🧬🌐

Kudos to Andrea, @dzanc.bsky.social & Sebastiano!

This is the world we have built over the last century.

But is it the world we deserve or that we should aim to change?
____
Data source: www.epi.org/publication/...

Forgot the lens: Nikkor 70-300

Nikon D750.
No filters.
No telescope.
Skyguider pro.

😂

Could humans and AI become a new evolutionary individual? | PNAS Could humans and AI become a new evolutionary individual?

Through recursive feedback, where humans shape AI, and AI increasingly shapes human thought and action, AI may acquire a role not as a separate agent, but as a core architectural element of an emerging collective individual

#ComplexityThoughts

www.pnas.org/doi/10.1073/...

Reduction down to a single source is abstracting away too much #information and losing the explanatory power of understanding deeper, #complex connections in an interdependent world.

The problem is, that this is the perfect way to mislead people w/BS, intentional or not.

bsky.app/profile/manl...

Screenshot for the many hidden worlds of quantum mechanics -- just the title and some evocative spheres floating around

"The Many Hidden Worlds of Quantum Mechanics," my lecture series for the Great Courses (24 half-hour lectures), is still available on Prime video for a few days! Tell your family that watching football on Thanksgiving is hopelessly uncool.

www.amazon.com/gp/video/det...

Working in theory/data-driven social science/methods for computational social science?

The second Italian Conference on Computational Social Science is coming: follow @cs2italy.bsky.social for updates and visit our website cs2italy.org for details.

🚨 Abstract submission deadline: January 15, 2026

Can a single number tell a story about a complex society? “Lies, damned lies, and statistics”

Excess mortality is *not* a simple count. It's a deviation capturing direct deaths, indirect effects, misclassification, delayed care: all shocks moving through a connected system. One number, many mechanisms.

w/ @adamjkucharski.bsky.social

manlius.substack.com/p/can-a-sing...

#ComplexityThoughts

Working in theory/data-driven social science/methods for computational social science?

The second Italian Conference on Computational Social Science is coming: follow @cs2italy.bsky.social for updates and visit our website cs2italy.org for details.

🚨 Abstract submission deadline: January 15, 2026

There is value in these thoughts over time.

16y ago was needed to argue for personalized medicine, w/ the preventive/predictive goals.

Nowadays we are experiencing personalized medicine.

What's missing is accurate models to achieve the P/P goals (ie, not just AI-driven, but mechanism-driven).

This is my personal taste, but most of the time I learn something quantitatively when I see equations 🫠

(predictive) AND (preventive) AND (medicine) - Search Results - PubMed (predictive) AND (preventive) AND (medicine) - Search Results - PubMed

“predictive” and “preventive” are widely used terms in the medical area. A quick search on pubmed shows that's the goal of much medicine:

pubmed.ncbi.nlm.nih.gov?term=%28pred...

I disagree that it “sounds similar”: maybe the purpose is similar (curious that we are still there), but the implementation is rather different.

That review does not propose a quantitative approach, we do.

And of course, with respect to 16 years ago, we have learned something!

Network Medicine is entering a new phase: one that demands we rethink how we study, model and ultimately treat complex diseases.

www.sciencedirect.com/science/arti...

#ComplexSystems #NetworkScience #Medicine #MedSky 🧪🧬🌐

🧵 1/

Eye lens radiocarbon reveals centuries of longevity in the Greenland shark (Somniosus microcephalus) Greenland sharks can live to be 400 years old and only become sexually mature at 150, raising conservation concerns.

400 years old shark.

Slow growth, slow metabolism.
No meetings, and no AI to enhance productivity.

www.science.org/doi/10.1126/...

Network Medicine 2.0 is not just a conceptual evolution: it is a necessary step to translate complex data and analytical tools into actionable clinical insights.

Working at the interface of network science, biology, machine learning or precision medicine?
👉 I’d love to hear your thoughts.

/fin

⁉️ Why this matters at all?

A rigorous, multiscale, uncertainty-aware framework to:

🧬 understand disease mechanisms
🧪 discover biomarkers
💊 identify targets & repurpose drugs
🧠 interpret single-cell & spatial omics
🏥 design predictive, preventive & personalized medicine (e.g., #DigitalTwins)

7/

3️⃣ Mechanistic models anchored in physics and chemistry: not (just) abstract topology

Network structure alone is not (and cannot be) enough. Incorporating physicochemical constraints — binding affinities, conform. ensembles, .. — leads to real dynamics of living systems & their perturbations

6/

2️⃣ Uncertainty-aware reconstruction: not perfect networks.

From incomplete interactomes to noisy correlations, biological networks are measured with error.
We need probabilistic, generative models able to explicitly encode uncertainty, rather than hiding it.

5/

1️⃣ Interdependent, multilayer systems: not single networks

Biology operates across interacting layers: genome, proteome, metabolome, cell–cell communication, tissue-level dynamics. Multilayer modeling captures these cross-scale dependencies and gives us a more realistic view of disease processes

4/

TL;DR?

👉 Biological datasets are vast and heterogeneous: yet many models still treat systems as static, complete and noise-free, assumptions that simply don’t hold.

Therefore, we argue for a paradigm shift built on three pillars.

3/

🚨 In our new paper led by @valedand.bsky.social in collaboration w/ Prof. Joe Loscalzo harvardmed.bsky.social, just published in @cp-med.bsky.social, we outline why the next generation #NetworkMedicine 2.0 must move beyond isolated system components to embrace true biological #complexity

2/

Network Medicine is entering a new phase: one that demands we rethink how we study, model and ultimately treat complex diseases.

www.sciencedirect.com/science/arti...

#ComplexSystems #NetworkScience #Medicine #MedSky 🧪🧬🌐

🧵 1/

It'd be lovely to see the experiment.

Unexpected events and prosocial behavior: the Batman effect - npj Mental Health Research npj Mental Health Research - Unexpected events and prosocial behavior: the Batman effect

This study tested whether an unexpected event, such as the presence of a person dressed as Batman, could increase prosocial behavior by disrupting routine.

Passengers were significantly more likely to offer their seat when Batman was present.

www.nature.com/articles/s44...

What a location for a seminar.👀

#ComplexSystems #Padova

Image of the Pleiades, in a light blue glow

Not perfect 📸, but fair enough for 1h exposure.

Bortle 7 sky.
No telescope, no AI.

#Astrophotography #Pleiades #M45