Of course that's what they do, because it makes them look like the good guys, who do the right thing first. Then the onus of effort is on the people who come afterwards, even if they're the ones who are in fact solid. Disgusting, but unsurprising.
The book is available through Amazon (t.co/9qvCkYKeLJ) or on the publisher website (t.co/agmYYi66IW). Many universities got it through subscription, so you can check here onlinelibrary.wiley.com/doi/book/10.... or with your library for free access.
Thanks for sharing!
I wrote this together with Prof. David Eisner, and it was honestly one of the smoothest and nicest collaborations ever. You can find more about the background of the project and how it relates to punk rock here: t.co/seVYPxAth6
5/
It will help you figure out which statistical test to use for what question and data, and what to be careful about when using them. How to avoid common issues such as pseudoreplication and multiple testing problem. We also cover basics of data visualization and experimental design.
4/
Existing books are typically too long, too complex, too mathsy, and at the same time often don't go into areas that essential for scientists.
We took an "maximum understanding, minimum maths" approach, fitting a lot of relevant content in under 200 pages.
3/
Statistics is a key element of today's science. Unfortunately, its ubiquity is parallelled by ubiquity of problematic use, which negatively affects interpretation of results and reproducibility.
Key issue was - how to learn practical statistics for life sciences?
2/
Author copies of our new book on statistics (mainly) for life scientists arrived recently. After three years of work, this is exciting, especially as we really hope it can help make a positive difference.
What's so special about it?
π§΅
Thanks so much to everyone involved, including Xin Zhou, Hector Martinez-Navarro, Maxx Holmes, Lucas Arantes Berg, Tom Bury, Marketa Tomkova, Emily Jo, Norbert Nagy, Ambre Bertrand, Alfonso Bueno-Orovio, @physicsoftheheart.bsky.social, Don Bers, Blanca Rodriguez, and Jordi Heijman!
Finally, if you want to study how a signalling pathway, a drug, or a nonstandard channel affect physiology, model generality is key for obtaining a complete picture of the effects and benefits/risks.
Happy to answer any questions/discuss suitability for your projects!
Second, real-world arrhythmia arises from interactions of arrhythmia triggers (including afterdepolarisations) and substrate (alternans, steep S1-S2 restitution). To reconstruct different origins of arrhythmia, we need a single model capable of all of those. 8/X
First, it is about general credibility. Models are, like real cells, full of interconnections and feedback loops. This means if a model has a mechanistic problem in one component, this can propagate into other areas of use and cause incorrect predictions. 7/X
T-World will be made freely available as open-source code (Matlab, CellML, C, CUDA) and online graphical user interface to facilitate use by non-coders.
As a final note, why does generality of a model matter so much?
6/X
How did we achieve this? The name of the model reflects contribution of labs from the whole world over last decades. We integrate ideas from the most influential families of cardiac models (Rudy and Bers/Grandi) and numerous new ideas and improvements. 5/X
T-World demonstrates strong performance in predicting drug-induced arrhythmia risk and opens new opportunities for predicting and explaining drug efficacy, demonstrated by unpicking effects of mexiletine in Long QT syndrome 2. 4/X
Using T-World we revealed a likely sex-specific arrhythmia risk in females related to restitution properties, identified arrhythmia drivers in type 2 diabetes, and describe unexpectedly direct possible pro-arrhythmic role of NaV1.8 in heart failure. 3/X
T-World recapitulates all key cellular drivers of arrhythmia: early and delayed afterdepolarisations, alternans, and steep S1-S2 restitution. We can now study their interactions in tissue, or carry out multifactorial assessment of drugs (including contractility). 2/X
After seven years of development, we present the highly general human ventricular cardiomyocyte T-World www.biorxiv.org/content/10.1...
It represents excitation-contraction coupling, contractility, sympathetic signalling, sex differences, and it is suitable for multiscale studies. 1/X
Phew, this was close, just checked and this was the one before...
OK!
(using this as a last slide in talks, with "THE END" added; sadly, fewer and fewer people get the reference)
Regression difficulty, ranked:
Easy: Linear regression
Medium: Nonlinear regression
Hard: Mixed-effects regression
Nightmare: Sleep regression
Pro tip for reference managers. If you are using Zotero to change in-text references' style from Author-year to numbered, it is not working no matter what, and you're getting desperate...
... do make sure it's not a text where you previously inserted references using Mendeley. π€¦ββοΈ
Was reading a book randomly and found a 14th century depiction of peer review...
Former postdoc Jakub Tomek has been awarded by the European Commission for his work on a computational model of a human ventricular myocyte (ToR-Ord) as part of Prof Blanca Rodriguezβs Computational Cardiovascular Science Team. www.dpag.ox.ac.uk/news/jakub-t...
@jakubtomek.bsky.social @ec.europa.eu
(from @pudukhepa.bsky.social )
Yes, it's often called epsilon in calculus and is quite a key concept!
I'd love that. Both computationally and experimentally, it would be great. I think the need for independence and niche at all costs preventing this is a bit silly and is not good for people, nor science.
Ten traktor je navΓc vynikajΓcΓ v tom, ΕΎe ten komΓn s trochou fantazie vypadΓ‘ jako lahev vΓna. Pro KDU ideΓ‘l.
The feeling before the next round of grant applications...
During my Oxford PhD, I was ok living on stipend that was much lower than what UK students labelled as "basically unlivable". It's a lot about expectations, what's considered a "basic standard", and what one finds rewarding in life. 4/4
