iiinova — where your digital self keeps evolving

Iain Johnston @mitomaths.bsky.social · 04/09/2026

... finding striking similarities of evol dynamics across different pathogens, but also notable diffs predicted by geography and patient demographics. The networks (grey) are predictive, and perform well forecasting new resistome phenotypes (coloured dots) from sub-Saharan Africa... 2/

110

Iain Johnston @mitomaths.bsky.social · 04/09/2026

How does AMR evolution vary across pathogens? Regions? Patients? We used EvAM* to take a look. www.biorxiv.org/content/10.6... Inferring evol pathways to multi-drug resistance as transition networks, we use the amazing CABBAGE dataset to learn AMR "roadmaps" across ESKAPEEs around the world... 1/

220

Iain Johnston @mitomaths.bsky.social · 30/06/2026

Super grateful to Dr Robert Glastad (and maker @neurowoodworks.bsky.social ) for this beautiful art celebrating Robert's successful PhD defence! Featuring Eqn. 15 journals.plos.org/ploscompbiol... linking physical behaviour of mitochondria and genetic behaviour of mtDNA :-)

020

Iain Johnston @mitomaths.bsky.social · 09/06/2026

We apply it to the evolution of drug resistance in TB and Klebsiella, and show its interoperability with other problems. The link is www.biorxiv.org/content/10.6... ... would love to hear thoughts!

000

Iain Johnston @mitomaths.bsky.social · 09/06/2026

This sets up for a version of the Mk model that scales linearly in observations instead of exponentially in features, and can handle interactions, reversibility, coupling, missing data, and more. We think this could be quite a generally useful advance! 4/

101

Iain Johnston @mitomaths.bsky.social · 09/06/2026

One can -- the classic Mk model in evol bio. But it struggles with many interacting features, because the possible state space becomes huge. But, though possible, many of these states are unlikely. So, basically, we prune them in a rational way. 3/

120

Iain Johnston @mitomaths.bsky.social · 09/06/2026

Excited to share a new preprint! Lots of evolutionary processes, from AMR to cancer, involve biology acquiring features (drug resistances, mutations, etc). These are sometimes called "accumulation pathways", an instance of discrete character evolution. 1/

140

Iain Johnston @mitomaths.bsky.social · 02/04/2026

Mito evolution special issue of Phil Trans now out! royalsocietypublishing.org/rstb/issue/3... Excited to be part of this, with some new perspectives on mtDNA segregation from us royalsocietypublishing.org/rstb/article... and brilliant content from colleagues worldwide! Pics from current location 🇳🇴

051

Iain Johnston @mitomaths.bsky.social · 28/02/2026

Vinterferie ❤️🇳🇴🦢

020

Iain Johnston @mitomaths.bsky.social · 09/01/2026

1. Apply for one of the positions below (lab or theory) by Jan 18th 2. Learn interdisc skills and discover cool new things about how mitochondria move and socialise 3. Explore some of Norway's beautiful nature (both the below, Rundemanen and Gullfjellet, <10km from work)

058

Iain Johnston @mitomaths.bsky.social · 18/12/2025

Mitochondria, hypercubes, and a blackboard -- what a lot of favourite things for J Evol Biol this month! Thanks to all involved for a smooth and pleasant review editorial process; academic.oup.com/jeb/article/...

050

Iain Johnston @mitomaths.bsky.social · 15/11/2025

Paradise up on the mountain trails

020

Iain Johnston @mitomaths.bsky.social · 22/09/2025

Understanding evolutionary dynamics lets us form future predictions. We tested these predictions with newly sequenced data from decades of clinical study in Tanzania. Fitted model retrospectively predicted those previously unobserved accumulation dynamics! 4/

110

Iain Johnston @mitomaths.bsky.social · 22/09/2025

Interesting, but very high-dimensional, so we project down to find the principal axes of diversity in evolutionary behaviour -- and what predicts them. They're linked both to public health region and drug use regimens 3/

100

Iain Johnston @mitomaths.bsky.social · 22/09/2025

Following the ideas in journals.asm.org/doi/full/10.... about applying evol accumulation modelling (EvAM) to AMR, we use genomics data to infer the orderings of character accumulation behind observed Klebsiella AMR profiles across countries (e.g. Romania here), with global average in D 2/

100

Iain Johnston @mitomaths.bsky.social · 22/09/2025

Great to see this out there! We infer the evolutionary pathways behind patterns of AMR in Klebsiella across 102 states, finding globally consistent principles, state-specific deviations (and covariates), and validating evol predictions with new genomes. biorxiv.org/content/10.1... 1/

1114

Iain Johnston @mitomaths.bsky.social · 17/06/2025

Celebrating grant award with a mountain run! MitoPhyto, a FRIPRO project, will combine modelling and microscopy to explore mitochondrial and mtDNA maintenance across non-bilaterian multicellular eukaryotes. Opportunities to come to Norway: please see next post

111

Iain Johnston @mitomaths.bsky.social · 17/05/2025

Gratulerer med dagen alle sammen! (non-Norway-based folks, happy Norwegian Constitution Day!)

000

Iain Johnston @mitomaths.bsky.social · 07/05/2025

Had a brilliant time at the 54th Jírovec’s Protozoological Days in Frymburk. Amazing organisms, amazing science, lots of exciting organelle work. Beautiful setting too! Thanks to all www.chromera.org/konference/5...

000

Iain Johnston @mitomaths.bsky.social · 07/05/2025

May trails on the way to the final Applied Statistics lecture of the semester!

020

Iain Johnston @mitomaths.bsky.social · 02/05/2025

View from the office as summer approaches

020

Iain Johnston @mitomaths.bsky.social · 19/03/2025

Projector broke today, so a good old-school blackboard lecture on an intro to GLMs!

010

Iain Johnston @mitomaths.bsky.social · 26/02/2025

Started likelihood-based approaches for model selection in STAT200 today, one of my favourite things (and behind many of our research projects)! Combined with likelihood-free methods, my claim is that this is how we turn modelling (not really falsifiable) into science (testing hypotheses with data)

000

Iain Johnston @mitomaths.bsky.social · 20/02/2025

Some shots from wintery commutes in 2025!

010

Iain Johnston @mitomaths.bsky.social · 20/02/2025

Nice group meeting today, celebrating exciting new positions and discussing the diverse ways we do science with data! Thanks all

031

Iain Johnston @mitomaths.bsky.social · 20/02/2025

And very new indeed, here's HyperLAU, an approach using tools from linear algebra to capture uncertainty in (phylogenetically embedded) accumulation models that supports flexible model structure and data inputs @jessirenz.bsky.social ! arxiv.org/abs/2502.05872 . Would love to chat about these! 3/3

001

Iain Johnston @mitomaths.bsky.social · 20/02/2025

Here's a "picture guide" (led by Ramon) describing approaches for these models and their inference, from cancer to evol bio and other applications arxiv.org/abs/2312.06824 (also see Ramon and colleagues' great repo of software approaches for EvAM github.com/rdiaz02/EvAM...) 2/

101

Iain Johnston @mitomaths.bsky.social · 20/02/2025

We've been thinking a lot about "accumulation models" lately -- inferring how systems accumulate features over time. May be mutations in a tumor, characters in evol bio, gene losses in oDNA. Here's an instance of the well-known Mk model that captures this picture academic.oup.com/bioinformati... 1/

101

Iain Johnston @mitomaths.bsky.social · 02/01/2025

First skitur of 2025, chasing the sunrise

030

Iain Johnston @mitomaths.bsky.social · 11/12/2024

More beautiful commuting on the trails, featuring dramatic mist and Norway's biggest tree, celebrated in typical understated style 🇳🇴🌲

020

Iain Johnston @mitomaths.bsky.social · 11/12/2024

Here's a mitochondrial "social network" (tracking encounters) emerging over time from their collective dynamics in a plant cell. This picture helps us understand their capacity to share and exchange biomolecules 🔬🌱🫂 Read more www.sciencedirect.com/science/arti...; pubmed.ncbi.nlm.nih.gov/34015261/

1194

Iain Johnston @mitomaths.bsky.social · 10/12/2024

First commute run for a while, following some illness, travel, and language exams. Beautiful wintery vibes throughout over the Bergen mountains!

040

Iain Johnston @mitomaths.bsky.social · 09/12/2024

Have had some good chats recently about pivoting to non-profit and society journals. A point to share/underline is that such options exist across the "prestige" spectrum. Some examples in the image. All heavily subjective of course; would love to hear thoughts from others!

1013158

Iain Johnston @mitomaths.bsky.social · 06/12/2024

What's the biggest sofa you can move round a right-angled corridor one metre wide? One of my favourite "easy to ask, hard to answer" maths problems has a [claimed] solution after many, many decades (!) of work arxiv.org/abs/2411.19826 (GIF from blogs.ams.org/visualinsigh...)

131

Iain Johnston @mitomaths.bsky.social · 29/11/2024

Wonderful time visiting and talking at Oslo Natural History Museum www.nhm.uio.no ! A childhood dream come true to present science in such a cool environment. Looking forward to more exploring today!

030

Iain Johnston @mitomaths.bsky.social · 29/11/2024

EvAM approaches can learn the ordering of resistance acquisitionsᵇ from dataᵃ -- but also which resistance features influence which othersᵈ, what will likely happen in futureᶜ, likely values of unobserved features, differences in evolution between regionsᵉ, and more. 3/

110

Iain Johnston @mitomaths.bsky.social · 29/11/2024

RTs much appreciated... would love to reach both AMR and biomath folks! "Evolutionary accumulation modelling in AMR: machine learning to infer and predict evolutionary dynamics of multi-drug resistance" arxiv.org/abs/2411.00219 We think EvAM methods have some potential in AMR. 🧵

1105

Iain Johnston @mitomaths.bsky.social · 10/07/2024

(3-4/5) Extents of chloroplast and mitochondrial genome reductions are correlated across diverse species academic.oup.com/mbe/article/... and features predicting cp/mt gene retention also appear in profiles of more recent endosymbioses (incl nitroplast) academic.oup.com/gbe/article/...

030

Iain Johnston @mitomaths.bsky.social · 10/07/2024

(2/5) ATP variability influences cell fate decisions by modulating dynamics of gene regulatory networks royalsocietypublishing.org/doi/10.1098/...

000

Iain Johnston @mitomaths.bsky.social · 10/07/2024

I've neglected this space for a while and am reminded of its relative strengths on a revisit! Here are some recent papers that may be of interest: (1/5) Sea anemones grow and shrink over orders of magnitude with/without food journals.biologists.com/dev/article/...

010

Iain Johnston @mitomaths.bsky.social · 21/12/2023

Thanks @newphyt.bsky.social for the cover choice! The pic -- a family collaboration! -- shows heteroplasmic cell mixtures of organelle DNA (green and white) in the plant's early stages "segregating" over time towards homoplasmy (green or white). nph.onlinelibrary.wiley.com/doi/10.1111/...

051

Iain Johnston @mitomaths.bsky.social · 18/12/2023

As an early Christmas present to HyperHMM academic.oup.com/bioinformati... we gave it the Rcpp link it's wanted all year! Even easier to learn accumulation pathways in evolution (incl AMR) and cancer with unrestricted dependencies between features. github.com/StochasticBi...

000

Iain Johnston @mitomaths.bsky.social · 15/12/2023

We also looked at conditional dependencies between losses of specific mt and cp genes. A few mt genes are inferred to necessarily precede the loss of a few cp genes, but the two genomes overall display little conditional dependency (with limited data) 3/

100

Iain Johnston @mitomaths.bsky.social · 15/12/2023

There's a more limited set of organelle genomes available for organisms that have both mtDNA and cpDNA. But taking these (A) and attempting to control for relatedness (B) we do find evidence for an mt-cp correlation (albeit weak). 2/

110

Iain Johnston @mitomaths.bsky.social · 15/12/2023

A while back we found that the same features predict gene loss from mitochondrial and chloroplast DNA across eukaryotes cell.com/cell-systems... . But that left an open Q -- is mtDNA and cpDNA loss linked at the level of individual species? New preprint: biorxiv.org/content/10.1... 1/

120

Iain Johnston @mitomaths.bsky.social · 30/11/2023

While discussing the tensions and resolutions that may shape mito dynamics, we also highlight the global "One Health" importance of plant mitos and mtDNA in producing food. Thanks, as always, to reviewers and the editors for the invitation. Would love to hear thoughts!

000

Iain Johnston @mitomaths.bsky.social · 30/11/2023

These conflicting priorities arise across kingdoms, and we focus on plants, both because they're a great model and because plant mitos feed the world -- both by powering plants and by facilitating hybrid production (paradoxically, often through their dysfunction and CMS).

100