Screen shot of a youtube video titled "AI.FILL Function Explained: 10X Productivity in Excel with AI" with the caption "Let ChatGPT fill your missing data"

Don't you f**king dare.

Pairwise MCMC posterior samples from inference of parameters in our Noble1962-with-concentrations model.

...which can be used in optimisation or inference schemes like MCMC where thousands of action potentials at different parameter values are required, reducing the computation time by up to 70%. Open source Julia code provided. 🧵6/6

(very nice copyediting in Physica D with marked up LaTeX changes 😀)

box and whisker plot showing compution times for large and small parameter perturbations, using continuation methods versus standard ODE solving.

..including a novel "Noble 1962 with added concentrations" to use as a toy model, but also the Dutta et al. (CiPA version of the O'Hara model) which is a modern complex ODE system. The result is substantially faster calculation of limit cycles under parameter perturbations... 🧵5/6

...BUT there is another way to get the limit cycle. Once you've computed one, you can use 'continuation methods' to explore how this solution changes as a function of parameters, in this paper we get this working for electrophysiology models for the first time (we think?)... 🧵4/6

...this is usually the most relevant physiological behaviour to fit to data. Simulating all the way to a relatively good numerical approximation of the limit cycle usually involves solving ODEs for thousands of paces (millions of timesteps). Wrapped in an optimiser? Billions!....🧵3/6

convergence of an action potential over time as a limit cycle in model state space

... when doing electrophysiology simulations there are 'slowly changing' variables - usually concentrations - that take hundreds or even thousands of paces to settle down to a steady response. In dynamical systems parlance this steady response is a stable periodic orbit or "limit cycle"....🧵2/6

🚨 Latest paper out now 🚨 by @mattjowen.bsky.social and me: "Continuation methods as a tool for parameter inference in models of biological oscillations: A case study in electrophysiology modeling" in
doi.org/10.1016/j.ph...
short tweetorial follows ..... 🧵1/6

...BUT there is another way to get the limit cycle. Once you've computed one, you can use 'continuation methods' to explore how this solution changes as a function of parameters, in this paper we get this working for electrophysiology models for the first time (we think?)... 🧵4/6

...this is usually the most relevant physiological behaviour to fit to data. Simulating all the way to a relatively good numerical approximation of the limit cycle usually involves solving ODEs for thousands of paces (millions of timesteps). Wrapped in an optimiser? Billions!....🧵3/6

... when doing electrophysiology simulations there are 'slowly changing' variables - usually concentrations - that take hundreds or even thousands of paces to settle down to a steady response. In dynamical systems parlance this steady response is a stable periodic orbit or "limit cycle"....🧵2/6

AI is normalising n=1 experiments – and that’s a problem Why your headline result may well just be a coin toss

New post, on how AI is normalising n=1 experiments: kucharski.substack.com/p/ai-has-an-...

I know, I know, I'm doing it the wrong way round

Me writing EU grant applications:

Part B1: I can't possibly squeeze everything they need to know in there.

Part B2: but I already explained exactly what you need to know precisely and accurately, I can't make it longer now

How can this go on for another 3 years?

Data Organization in Spreadsheets Karl W. Broman & Kara H. Woo Pages 2-10 | Received 01 Jun 2017, Accepted author version posted online: 29 Sep 2017, Published online: 24 Apr 2018 1. Introduction 2. Be Consistent 3. Choose Good Names for Things 4. Write Dates as YYYY-MM-DD 5. No Empty Cells 6. Put Just One Thing in a Cell 7. Make it a Rectangle 8. Create a Data Dictionary 9. No Calculations in the Raw Data Files 10. Do Not Use Font Color or Highlighting as Data 11. Make Backups 12. Use Data Validation to Avoid Errors 13. Save the Data in Plain Text Files ABSTRACT Spreadsheets are widely used software tools for data entry, storage, analysis, and visualization. Focusing on the data entry and storage aspects, this article offers practical recommendations for organizing spreadsheet data to reduce errors and ease later analyses. The basic principles are: be consistent, write dates like YYYY-MM-DD, do not leave any cells empty, put just one thing in a cell, organize the data as a single rectangle (with subjects as rows and variables as columns, and with a single header row), create a data dictionary, do not include calculations in the raw data files, do not use font color or highlighting as data, choose good names for things, make backups, use data validation to avoid data entry errors, and save the data in plain text files.

Every day is a good day for sharing one of the most useful papers about research data ever written. PLEASE get your people to understand and follow this advice.

www.tandfonline.com/doi/full/10....

So, BlueSky, one of my passions is the study of tea in WW2. I need you to watch this oral history from three ex-tea ladies from the Bermondsey WVS.

Because you could guess for ONE THOUSAND YEARS and you would not guess where it is going... #skystorians #history #tea

In the whole carriage maybe 12 people!

Peak time express train 6.02pm out of St Pancras. In the 12 seats around me there is one other person. Pricing must be significantly higher than optimal for (a) customers, (b) income for train company, (c) the wider economy. On the plus side it's very pleasant.

Explainer: UK government has announced a suspension on student visa applications from nationals of Afghanistan, Cameroon, Myanmar and Sudan. Details here www.gov.uk/guidance/vis...

Data

Pretty convinced that nursery fees being in the basket of goods is solely responsible for keeping inflation above 2%.

...<googled>...

Oh look, nursery fee inflation actually is 8% 🤯!

www.ons.gov.uk/economy/infl...

Not sure whether to be relieved ours isn't an outlier or to start a nursery.

(1) Read, and replace any mention of 'work' or 'papers' in your mind with 'words in order'.

(2) Repeat, replace the same with 'new knowledge'.

Which makes most sense?

If the article is about words-in-order, discard. If it is about new knowledge, read it carefully to see if the author is loopy.

It's never not been crazy-making to me that we'll have hundreds of thousands of words written about a 2% change to income tax either way and the impact it'll have in the economy, but 9% of half the population under 35 is waved away as a private matter, or "only the price of a gym membership".

Refugee status becomes temporary in asylum shake-up Adults and accompanied children claiming asylum will only receive refugee status for 30 months under the changes.

Abysmal, immoral, cruel, small hearted, tiny minded, ineffective, undemocratic, and just generally utterly indecent. What a bitter godforesaken disappointment these people are.
www.bbc.co.uk/news/article...

Strong Message Here - History Will Judge You (with Stewart Lee) - BBC Sounds Is history still written by the victors? How do people try and control their legacy?

New Ep now up. Stewart Lee and I discuss swearing at the BAFTA’s, Blair’s legacy and Gerry Adams trampolining naked with a dog. #StrongMessageHere www.bbc.co.uk/sounds/play/...

I'll just repost this today #ByElection #UKPolitics

The government's immigration plans: a dog’s dinner of half-conceived policy, Reform rhetoric, mean-spirited sentiment, and profound intellectual muddle iandunt.substack.com/p/can-we-ext...

A graph from the financial times showing that rule changes have increased mean student loan repayments from £40k to £56k

Substantial goal post moving since they opted in knowing the system:

(From "It’s time to phase out student loans" - www.ft.com/content/797c... via @financialtimes.com )

‘Is university still worth it?’ is the wrong question The graduate earnings premium isn’t really measuring what most people think

‘Is university still worth it?’ is the wrong question - giftarticle.ft.com/giftarticle/... via @FT

Agree it should be considered graduate tax. But only youngest people pay, after massive shift of wealth to older people via house prices. Plus v. rich opt out (by paying fees upfront). Progressive 🤨?

Imagine furore if 9% was added to everyone's higher rate tax. We'd be disincentivising work etc...