I couldn't find a tool to plot different #neuroimaging data in one consistent style, so I made one! Meet yabplot (yet another brain plot) - a #Python package for (sub)cortex & tracts.🧠
- Simple API
- Built-in atlases
- Custom atlas support
🔗 github.com/teanijarv/ya... (drop a ⭐️!)
🩸 🧠 ⏱️ What if a blood test could predict WHEN an individual would develop symptoms of Alzheimer’s disease? High plasma p-tau217 predicts greater risk for Alzheimer’s symptoms—do the plasma p-tau217 levels provide insights into WHEN symptoms might begin?
nature.com/articles/s41...
By capturing the balance between aggregate formation and cellular clearance, our model explains decades of stability before sudden runaway dynamics, and offers a framework to predict disease onset and therapeutic efficacy. Mostly driven by Matthew Cotton and Georg Meisl doi.org/10.1063/5.03...
🚨PREPRINT
In our new paper, we link RNNs and neural manifolds by introducing the DDM framework.
We can train networks to embed an arbitary dynamical system in a latent subspace. We illustrate this with simple models of input-driven and autonomous associative memory. Enjoy!
arxiv.org/abs/2602.14885
Huge thanks to all my coauthors @jwvogel.bsky.social, Travis Thompson, Roxana Aldea, Olof Strandberg, Erik Stomrud, Sebastian Palmqvist, @rikossenkoppele.bsky.social, @saadjbabdi.bsky.social, Stefano Magon, Gregory Klein!, Niklas Mattsson-Carlgren, Oskar Hansson, and @alaingoriely.bsky.social
16/16
Overall, dATN unifies mechanistic biology with clinical imaging biomarkers to:
✔ forecast disease progression
✔ test mechanistic hypotheses
✔ optimise personalised treatment strategies in AD
For all the details, please read the preprint! And do share your thoughts with us!
15/
We show that intervention with Aβ-targeting therapies before Aβ/tau colocalisation yields greater benefits on tau and neurodegeneration, and diminishing benefits after colocalisation. This provides a mechanistic explanation for why early intervention may be important for therapeutic success.
14/
...and here's what it looks like with Aβ-targeting intervention. We can spatially simulate changes in ATN biomarkers in response to treatment!
13/
Here's a simulation showing ATN progression with a rescaled dATN model (everything evolves between 0 and 1) without Aβ-targeting intervention.
12/
Finally, we demonstrate how the dATN model can be used to enhance pharmacokinetic-pharmacodynamic simulations of AD through regional mechanistic modelling. We show how we can evaluate Aβ-targeting therapies spatially and identify mechanism-based critical intervention windows.
11/
We show that the spatial overlap of Aβ and tau is a critical biomarker of disease acceleration, not just their global burden. The colocalisation point demarcates the transition from transport-dominant tau spread to Aβ-induced production-dominated cortical tau spread (caTAUstrophy!)
10/
Next, we looked at when and where Aβ and tau colocalise. Using the dATN model and learned parameters, we can simulate the full colocalisation order. We show that initial colocalisation in the inferior temporal lobe occurs just before widespread cortical colocalisation in known Aβ-rich tau-hubs.
9/
The inferred parameters show tau transport and production are of similar magnitude. Tau becomes production-dominated only when interacting with Aβ, highlighting a mechanistic role for Aβ in accelerating tau pathology.
8/
We calibrated dATN using longitudinal multim
odal imaging from A+T+ individuals in two cohorts:
📊 ADNI
📊 BioFINDER-2
The model accurately fits regional Aβ, tau, and neurodegeneration trajectories.
7/
The first key insight is that cortical Braak-like staging of tau can emerge naturally from network-based spread combined with local Aβ-driven tau production, without imposing staging by design.
6/
The result is a simulatable, personalisable, mechanism-based model of regional Aβ, tau, and neurodegeneration progression across the brain.
Here, we show a simulation with a small dense Aβ seeding, tau seeds in bilateral entorhinal cortex, and zero baseline atrophy.
5/
To address this, we introduce the dynamical ATN (dATN) model. It mechanistically simulates the spatiotemporal evolution of Aβ, tau, and neurodegeneration via:
• Prion-like aggregation
• Network tau spread
• Aβ-driven tau production
• Tau-driven atrophy
4/
The ATN and revised NIA frameworks classify disease stages using biomarkers—but they are largely descriptive and do not predict how Aβ, tau, and neurodegeneration interact along the AD continuum.
3/
Alzheimer’s disease is shaped by interactions between Aβ and tau, which together drive neurodegeneration. However, how these processes unfold across brain regions and over time remains unclear.
2/
‼️ New preprint ‼️
How do amyloid-β (Aβ) and tau drive Alzheimer’s disease over time?
We introduce a parsimonious, mechanism-based dynamical ATN (dATN) model to simulate longitudinal imaging biomarkers. A short thread 👇
Preprint: www.biorxiv.org/content/10.6...
❗🧠 New review in IEEE RBME
How network math models are reshaping how we think about neurodegenerative disease, from brain dynamics to disease progression
"Network models of neurodegeneration: bridging neuronal dynamics and disease progression"
ieeexplore.ieee.org/document/113...
🚨PREPRINT
Coarse-graining diffusions with discrete-state approximations is an easy and effective way to build stochastic models from observed trajectories, but how valid is this approximation?
arxiv.org/pdf/2511.05366
We focus on this problem, with an emphasis on the nonequilibrium steady-state.
Our review paper on nonequilibrium physics in the brain is now out in Physics Reports!
doi.org/10.1016/j.ph...
How does the 🧠 process information?
In this review we show how to use the tools of non-equilibrium thermodynamics to understand brain dynamics in discrete and continuous state spaces
Great work led by Ramon as part of his PhD
kwnsfk27.r.eu-west-1.awstrack.me/L0/https:%2F...
does brain connectivity drive spread of pathological proteins in Alzheimer’s disease?
✨ preprint: www.biorxiv.org/content/10.1...
if the question intrigues you, please read on 🧵⤵️
For more details, please check out the paper! A huge thanks to my coauthors @alaingoriely.bsky.social @jwvogel.bsky.social @alexapb.bsky.social @saadjbabdi.bsky.social @biofinder.bsky.social (and others not on Bluesky) for helping to get this over the line.
With help from Oskar Hansson, @jwvogel.bsky.social, @alexapb.bsky.social and the @biofinder.bsky.social team, we were also able to replicate this result in an independent dataset with a different tau tracer, the BF2 dataset!
We then fit the model to biomarker groups across the AD continuum in ADNI. Our results show an initial transport-dominated phase (A+T-), followed by a production-dominated phase (A+T+). Suggesting early cortical tau deposition via network diffusion is followed by accelerated local tau production.
We used a series of test/train splits on A+T+ ADNI subjects with at least 4 scans to show how adding data changes prediction uncertainty. We show that too few scans can lead to misleading predictions, but with 3 longitudinal scans, regional predictions are pretty good! (here showing left inf temp.)
Through model comparison on an A+T+ group from ADNI, using in-sample and out-of-sample measures, we show that regional heterogeneities are essential for longitudinal modelling of tau progression and highlight deficiencies in a network diffusion model and models with homogenous production dynamics.
