Gary Mirams
@mirams.bsky.social
Prof. of Mathematical Physiology, University of Nottingham. Like: rambling, scrambling, climbing, surfing, Irish trad music, VW vans.
To be fair, they thought they'd managed to get it enacted in 2015, but Theresa May shelved it!
Absolutely agree it should have been top priority for UUK and every VC since then...
researchbriefings.files.parliament.uk/documents/SN...
Absolutely agree it should have been top priority for UUK and every VC since then...
researchbriefings.files.parliament.uk/documents/SN...
October 20, 2025 at 10:12 PM
To be fair, they thought they'd managed to get it enacted in 2015, but Theresa May shelved it!
Absolutely agree it should have been top priority for UUK and every VC since then...
researchbriefings.files.parliament.uk/documents/SN...
Absolutely agree it should have been top priority for UUK and every VC since then...
researchbriefings.files.parliament.uk/documents/SN...
🚨 We've written a #primer on #identifiability
Preprint: doi.org/10.48550/arX...
Brief to brilliant co-authors was "everything a PhD student needs to know about identifiability before trying to fit a model"
Fig shows how Fisher Info. becomes excellent with lots of data, even for nonlinear models.
Preprint: doi.org/10.48550/arX...
Brief to brilliant co-authors was "everything a PhD student needs to know about identifiability before trying to fit a model"
Fig shows how Fisher Info. becomes excellent with lots of data, even for nonlinear models.
September 24, 2025 at 12:23 PM
🚨 We've written a #primer on #identifiability
Preprint: doi.org/10.48550/arX...
Brief to brilliant co-authors was "everything a PhD student needs to know about identifiability before trying to fit a model"
Fig shows how Fisher Info. becomes excellent with lots of data, even for nonlinear models.
Preprint: doi.org/10.48550/arX...
Brief to brilliant co-authors was "everything a PhD student needs to know about identifiability before trying to fit a model"
Fig shows how Fisher Info. becomes excellent with lots of data, even for nonlinear models.
In #Sheffield today for #FickleHeart2025 meeting on uncertainty quantification in cardiac modelling and simulation
September 8, 2025 at 7:47 AM
In #Sheffield today for #FickleHeart2025 meeting on uncertainty quantification in cardiac modelling and simulation
Enjoying some time up on the West coast of the Highlands.
Despite its spectacular situation, and rugged and interesting nature, this is one of those 700m+ ridges ⛰️ that hardly anyone climbs because it's surrounded by Munros and Corbetts.
Quite the evening outing. Anyone guess where!?
Despite its spectacular situation, and rugged and interesting nature, this is one of those 700m+ ridges ⛰️ that hardly anyone climbs because it's surrounded by Munros and Corbetts.
Quite the evening outing. Anyone guess where!?
August 22, 2025 at 9:12 PM
Enjoying some time up on the West coast of the Highlands.
Despite its spectacular situation, and rugged and interesting nature, this is one of those 700m+ ridges ⛰️ that hardly anyone climbs because it's surrounded by Munros and Corbetts.
Quite the evening outing. Anyone guess where!?
Despite its spectacular situation, and rugged and interesting nature, this is one of those 700m+ ridges ⛰️ that hardly anyone climbs because it's surrounded by Munros and Corbetts.
Quite the evening outing. Anyone guess where!?
Just having a warm up
June 28, 2025 at 5:38 PM
Just having a warm up
Therein lies a thoroughly unpleasant tale
June 21, 2025 at 1:17 PM
Therein lies a thoroughly unpleasant tale
March 25, 2025 at 4:52 PM
This is a very Financial Times take on the success of weight-loss drugs.
They are so good, we need to get worried about pension life expectancy forecasts!
They are so good, we need to get worried about pension life expectancy forecasts!
March 24, 2025 at 1:22 PM
This is a very Financial Times take on the success of weight-loss drugs.
They are so good, we need to get worried about pension life expectancy forecasts!
They are so good, we need to get worried about pension life expectancy forecasts!
Not that Mum and Dad are hoarders, but they have got Queen Victoria's 1901 Bradford church memorial service pamphlet in their loft, for some reason.
March 16, 2025 at 6:19 PM
Not that Mum and Dad are hoarders, but they have got Queen Victoria's 1901 Bradford church memorial service pamphlet in their loft, for some reason.
I'm basically using #ChatGPT for one thing at the moment: making #matplotlib figure generation painless. Below was set up in about 20 minutes, 95% automatically, with a handful of prompts. Great for building up gradually: "now add 2nd panel showing", "now make it an animation by varying ..."
March 6, 2025 at 8:37 AM
I'm basically using #ChatGPT for one thing at the moment: making #matplotlib figure generation painless. Below was set up in about 20 minutes, 95% automatically, with a handful of prompts. Great for building up gradually: "now add 2nd panel showing", "now make it an animation by varying ..."
So how do mathematical action potential model predictions of APD prolongation compare with this? Here they are side-by-side (in the models we can just evaluate lots of combinations to get the heatmaps shown)....6/8
March 3, 2025 at 12:02 PM
So how do mathematical action potential model predictions of APD prolongation compare with this? Here they are side-by-side (in the models we can just evaluate lots of combinations to get the heatmaps shown)....6/8
So with Roche and Anabios we purposefully tested concs of compounds that blocked hERG and CaV simultaneously to examine how much APD prolongation we saw with somewhat 'balanced' blockers (also tested concs in samples w. mass spec). The resulting data looked like: 3/8
March 3, 2025 at 12:02 PM
So with Roche and Anabios we purposefully tested concs of compounds that blocked hERG and CaV simultaneously to examine how much APD prolongation we saw with somewhat 'balanced' blockers (also tested concs in samples w. mass spec). The resulting data looked like: 3/8
Here you go! (need to "\usepackage{amssymb}" )
January 26, 2025 at 10:42 AM
Here you go! (need to "\usepackage{amssymb}" )
What kind of #JunkMail do non-academics get these days? This is just over 24hrs worth of my #Junk folder. All is invites to contribute to journals, books, conferences, editorial boards.
Filter seems to do remarkable job of separating spam from genuine, 🤞because I won't be reading all this!
Filter seems to do remarkable job of separating spam from genuine, 🤞because I won't be reading all this!
January 23, 2025 at 2:36 PM
...but (there's always a BUT) you do have to be careful you get to the right* limit cycle, as it is possible to track to different places than you'd get running ODEs from same place...
*N.B. in cardiac modelling we aren't usually certain about initial conditions, so not really any more "right"! 9/9
*N.B. in cardiac modelling we aren't usually certain about initial conditions, so not really any more "right"! 9/9
January 17, 2025 at 11:48 AM
...but (there's always a BUT) you do have to be careful you get to the right* limit cycle, as it is possible to track to different places than you'd get running ODEs from same place...
*N.B. in cardiac modelling we aren't usually certain about initial conditions, so not really any more "right"! 9/9
*N.B. in cardiac modelling we aren't usually certain about initial conditions, so not really any more "right"! 9/9
70% reduction in runtimes is very possible, even for (relatively) simple ODE model. So do consider having a look at this kind of approach if your modelling problem / optimization / inference involves fitting limit cycles to data. ...8/9.
January 17, 2025 at 11:48 AM
70% reduction in runtimes is very possible, even for (relatively) simple ODE model. So do consider having a look at this kind of approach if your modelling problem / optimization / inference involves fitting limit cycles to data. ...8/9.
It's common in cardiac ephys (and lots of bio modelling situations) to want to fit oscillatory model outputs to data. Often the most sensible state to be fitting is the 'steady response' or 'limit cycle' in dynamical systems parlance. e.g. steady heart beat. ...2/9
January 17, 2025 at 11:48 AM
It's common in cardiac ephys (and lots of bio modelling situations) to want to fit oscillatory model outputs to data. Often the most sensible state to be fitting is the 'steady response' or 'limit cycle' in dynamical systems parlance. e.g. steady heart beat. ...2/9
Such a chilly start to day two of the Chaste Cell-Based Modelling workshop (chaste.github.io/workshops/20...) that the geese are standing on the lake
January 9, 2025 at 9:25 AM
Such a chilly start to day two of the Chaste Cell-Based Modelling workshop (chaste.github.io/workshops/20...) that the geese are standing on the lake
Are you @wellcometrust.bsky.social funded? I've updated my little flowchart on open access publishing under Wellcome policies to reflect changes coming into force on 1st Jan: #Wellcome #OpenAccess
December 10, 2024 at 11:42 AM
Are you @wellcometrust.bsky.social funded? I've updated my little flowchart on open access publishing under Wellcome policies to reflect changes coming into force on 1st Jan: #Wellcome #OpenAccess
Just completed our first year of solar panels + battery. Here's the monthly generation, import and export. Very nice to have a negative annual electricity bill!
November 29, 2024 at 10:47 AM
Just completed our first year of solar panels + battery. Here's the monthly generation, import and export. Very nice to have a negative annual electricity bill!