Cian O’Donnell / is on Nostr

Computational neuroscientist.
Lecturer in Data Analytics at Ulster University.
http://odonnellgroup.github.io

Last Notes

2024-09-23 15:01:22

Big fan of this plot showing ASD gene 'risk' on y-axis (i.e. the relative increase in diagnosis likelihood if you have a genetic variant) vs variant prevalence on x-axis (fraction of general population with that genetic variant).
Some high-risk genes that are rare (top left) and other low-risk genes that are common in population.
Common phenomenon seen in many disorders: high impact mutations don't often get propagated to the next generation.
From Willsey et al 2022: https://www.nature.com/articles/s41583-022-00576-7
https://static.mstdn.science/media_attachments/files/113/187/447/393/338/023/original/e5416291b13aa386.png

2024-07-17 22:49:46

If the ultimate purpose of memory is to guide our actions in future, what is the point of episodic memory?
Why do we remember details of our past experiences?

2024-06-21 18:32:03

Our summer school on Comp Neuro, Neurotech & NeuroAI:
in-person deadline 15/8
online deadline 10/8
feat.
- Athena Akrami
- Chiara Bartolozzi
- Cian O’Donnell
- Damien Coyle
- Daniela Tropea
- Girijesh Prasad
- Javier Minguez
- KC Santosh
- KongFatt Wong-Lin
- Liam McDaid
- Madeleine Lowry
- Nikola Kasabov
- Saugat Bhattacharyya
- Shirin Dora
https://www.ulster.ac.uk/faculties/computing-engineering-and-the-built-environment/computing-engineering-intelligent-systems/isrc-cn3-summer-school

2023-09-07 11:49:09

- reply

@npub1skv…aky3 thanks, that makes sense if we want to treat at a molecular level. But the worry is that for many disorders with complicated genetics (eg ASD, schizophrenia), or extremely rare genetic conditions, we may not have the knowledge or tools or money needed to make those low-level interventions.
We wrote a review previously arguing that in those cases, degeneracy may open up the chance for higher-level eg circuit treatments
https://www.sciencedirect.com/science/article/pii/S0959438821000787

2023-09-07 06:40:28

- reply

@npub1skv…aky3 100% (as I argued in a recent grant app)... but we still need to overcome the obstacle of degeneracy to do so?

2023-09-01 21:11:15

- reply

@npub1skv…aky3 @npub1u2v…4999 very good point, explicit "cards on the table" movements seem progressive. Although I'm not sure I would want a system that makes it too difficult for outlier approaches either. We all know the danger of groupthink eg Alzheimer's debacle

2023-09-01 19:55:08

- reply

@npub1skv…aky3 @npub1u2v…4999 that seems very... right! I have no idea how to set up the incentivization system accordingly though?

2023-09-01 18:39:55

- reply

@npub1skv…aky3 very interesting read, thanks for sharing.
I'm not a psychologist so can't relate directly, but it does remind me of an aspect of neuroscience that has bugged more the longer I've been here: that much of what we do and get rewarded for is transient and performative: giving talks make a few people feel something for a half a day, papers make a splash that peters out after a few months, grants which are nothing more than money for salaries and finite consumables.

2023-08-09 09:38:18

Updated version of my review on Nonlinear slow-timescale factors in synaptic plasticity: https://arxiv.org/abs/2306.07801
#neuroscience
https://static.mstdn.science/media_attachments/files/110/858/983/200/495/608/original/e28a054da826c772.jpeg