Can AI reveal risk & co-pathology of multiple neurodegenerative diseases from a single blood sample? We explored AI-based diagnostic power on high rank plasma proteomics (N=17,170). www.medrxiv.org/content/10.1... #neuroskyence #neurosky #Alzheimer #compneuro #AI #datascience #neurology

Thank you so much! Yeah, the gradients seem to be key to so many things. Will be v interested to read your paper when out!

Massive thanks to the
@biofinder.bsky.social team, @nfranzme.bsky.social team and all co-authors for sharing their data and insights – I’ve learnt so much!

@rikossenkoppele.bsky.social @aitchbi.bsky.social @teanijarv.bsky.social @ted-satterthwaite.bsky.social
and all non-bsky people

#MedSky #neuroskyence #neurosky #alzsky #compneuro #MRI #neuroimaging #neurology

As my first PhD proj I’m v happy to finally share our findings with the world! It’s been a ride, and I am immensely grateful to the tireless @jwvogel.bsky.social for putting up with me + our team for all support: DeMON lab (@xiaoyucaly.bsky.social ‪@anlijuncn.bsky.social‬ and ppl not here)

10/

The manuscript was written using #QuartoPub and is reproducible along with figures from start to finish with synthetic data at: github.com/DeMONLab-BioFINDER/fc_changes_follow_gradients

(although it may require some initial massaging to get started)

9/

In sum, we reveal an elegant organizational logic behind functional brain responses to aging and AD – we show these changes to have a dynamic nature, tie to them to cognitive performance, and suggest them to be neural responses to cognitive strain

8/

Why does this happen?

We hypothesize that the reason for these divergent patterns may be due to the brain trying to compensate for compromised cognitive function

7/

Why does this matter?

Because these gradient-aligned FC alterations were also strongly related to cognitive function – independently of age or AD

They could be signals of cognitive strain

6/

Importantly, we see longitudinal increase in AD pathology leading to similar FC patterns

Using a sliding window approach in the longitudinal subsample, we also replicate the non-linear pattern seen above

5/

Much work has pointed to FC increasing in early AD only to later decline

Using non-linear modelling we show that gradient-alignment seems to follow a similar trajectory: alignment happens early in the disease only to later decline

4/

We found that age and AD systematically alter FC across the entire cortex

These changes aligned w axes of brain organization which reflect core principles of cortical function:

🔹 Age → executive–nonexecutive axis (Gradient 3)
🔹 AD → sensory–association axis (Gradient 1)

3/

Prior studies of FC in age and AD have reported both hyper- and hypo-connectivity across different regions and networks, sometimes in discrepancy with one another

We took a broader view and examined these changes from the lens of large scale functional organization (gradients)

2/

How does the brain reorganize in the context of age and AD?

Are functional changes in age and AD similar?

Are these changes dynamic across the age/AD spectrum?

We unpack these questions in a sample of N=973 with AD biomarkers

Preprint: tinyurl.com/ymv8s7h8

⬇️🧵

1/

‼️New preprint!
We are happy to share our latest work led by @teanijarv.bsky.social investigating why tau pathology in AD often accumulates more in one hemisphere of the brain than the other.

Check out 🔗https://biorxiv.org/content/10.1101/2025.04.15.648728v1 or dive into the details below👇

Really important and cool work by @xiaoyucaly.bsky.social! Congratulations!