iiinova — where your digital self keeps evolving

Luke Pilling @lcpilling.bsky.social · 14/06/2026

Fantastic talk by @denizturkmen.bsky.social at #ESHG2026 on her study of oral corticosteroids: cumulative dose, adverse outcomes, and genetic modifiers of risk

011

Luke Pilling @lcpilling.bsky.social · 13/06/2026

Lots of great @exeter.ac.uk researchers here #eshg20266 with talks and posters: check them out!

031

Luke Pilling @lcpilling.bsky.social · 24/02/2026

Niche R package with epidemiology tools anyone? 📦 Named it {yodr} (phonetically "Yoda") which I cannot believe no-one has done before to be honest! Main functions are for "tidy" model output (extends {broom} package) and PheWAS. Also functions for extreme p-values, etc lcpilling.github.io/yodr

yodr is a small R package with tools for epidemiology and hypothesis testing

020

Luke Pilling @lcpilling.bsky.social · 04/02/2026

My fav figure: shows the difference between full and partial genetic correlations for all 2485 disease pairs after adjustment for the genetics of BMI ⚖️ Blue shows where the genetics are reduced (explained) by BMI e.g., OA~T2D is fully attenuated 📉 Red shows where the genetic links are stronger!

Heatmap of the difference between full and partial genetic correlations for all 2485 disease pairs after adjustment for the genetics of BMI. Pairwise partial genetic correlation analysis between 71 conditions (i.e., 2485 disease pairs). Condition pairs where there was no significant (FDR correction) genetic correlation before or after adjustment, or the adjustment for BMI genetics was non-significant (FDR corrected), are coloured white. Data shown is the proportion change in genetic correlation between condition pairs after adjustment for BMI genetics (where the change is statistically significant after FDR correction). https://www.nature.com/articles/s43856-025-01347-y/figures/2

101

Luke Pilling @lcpilling.bsky.social · 19/01/2026

I don't normally do work on Saturdays (though often end up catching up in evenings due to young children etc) These Saturday GitHub commits were on nerdy personal repos where I annually analyse data about my central heating (Raspberry Pi managed) and my car fuel efficiency 😅 a wild Saturday evening

My 526 GitHub contributions (mostly commits) over the last 12 months. Never on Saturdays, rarely on Sundays, except where I have personal nerd projects apparently

030

Luke Pilling @lcpilling.bsky.social · 03/10/2025

One of the biggest increases in my analysis/coding productivity in recent years is completely embracing git repos 🐧 Coupled with GitHub it is *so easy* to be open, reproducible, efficient, and accountable to myself and colleagues 👨‍💻 No more chaotic servers wondering which script led to what! 😵‍💫

Cartoon by the excellent Allison Horst (https://allisonhorst.com/git-github). Includes a quote from Hadley Wickham and Jenny Bryan that says “Using a Git commit is like using anchors and other protection when climbing…if you make a mistake you can’t fall past the previous commit. Commits are also helpful to others, because they show your journey, not just the destination.” On the right, two little monsters climb a cliff face. Their ropes are secured by several anchors, each labeled “Commit”. Three monsters on the ground support the climbers.

2212

Luke Pilling @lcpilling.bsky.social · 16/09/2025

Post-doc opportunity in Exeter 🌟 studying genetic modifiers of haemochromatosis 🩸 Closing date 30 Oct! Come and work with a interdisciplinary team of epidemiologists, clinicians, and statistical geneticists, in a beautiful city in the South West of England ❤️ jobs.exeter.ac.uk/hrpr_webrecr...

View of the River Exe running through the Quay in sunny Exeter.

066

Luke Pilling @lcpilling.bsky.social · 03/09/2025

Latest preprint by PhD student Mitch Lucas on the genetic iron overload disease #haemochromatosis doi.org/10.1101/2025... Massive study of factors affecting clinical outcomes in HFE carriers, including genetic predisposition to higher transferrin saturation @janicelatkins.bsky.social @exeter.ac.uk

Cumulative incidence of haemochromatosis-associated clinical outcomes in males HFE C282Y homozygotes in UK Biobank, stratified into two groups based on transferrin saturation polygenic score (highest vs. lowest quintile). This shows clearly that even within the high-risk C282Y group, likelihood of adverse consequences is influenced by non-HFE genetic variants that also impact iron

030

Luke Pilling @lcpilling.bsky.social · 21/08/2025

Lots of instrumental music in my most listened artists... I cannot work if there is singing!

010

Luke Pilling @lcpilling.bsky.social · 21/08/2025

My latest data hobby: my Spotify listening history! Had no idea until recently that you could download a history over ever song you've ever listened to. Cue some fun R hours! Some surprising trends and top picks. Dominated by my work listening rather than "favourites" lcpilling.github.io/spotify

140

Luke Pilling @lcpilling.bsky.social · 09/07/2025

Exciting session on using the (appropriately named) "Epic" secondary care electronic healthcare records data for research by Prof Jane Masoli (pictured), Dr Nick kennedy and Michelle Ledbetter at the @swaih.bsky.social 2025 meeting @janemasoli.bsky.social @drnickkennedy.bsky.social

Professor Jane Masoli introducing how secondary care electronic healthcare records data (Epic) can be used for research

031

Luke Pilling @lcpilling.bsky.social · 16/06/2025

At the UK Human Functional Genomics Initiative @uk-fgx.bsky.social symposium @exeter.ac.uk, currently listening to intro from @jonathanmill.bsky.social Picked up my first fidget spinner from BioFAIR @biofair.bsky.social which sounds a really cool initiative for open data and methods 🌟

My new fidget spinner. I am holding it in the auditorium

010

Luke Pilling @lcpilling.bsky.social · 14/05/2025

First time my donated blood has made it to Swindon! 🤩 Love getting these e-mails of where my blood was used 🩸 Easiest way to save a life. Give blood! 🌟 www.blood.co.uk

Google maps showing the route from Exeter to Swindon via Bristol

030

Luke Pilling @lcpilling.bsky.social · 08/05/2025

Last session but certainly not least at multimorbidity MLTC symposium has reflections from stakeholders including one of our amazing public collaborators @marymanc.bsky.social This is my first project with such integrated patient perspectives and it has transformed the work. I'm really enjoying it!

Mary Mancini speaking on her experience as a patient with comorbodities

000

Luke Pilling @lcpilling.bsky.social · 08/05/2025

Next from GEMINI is Dr Olivia Murrin giving a deep dive on observational (from primary care) and genetic relationships between lung and musculoskeletal conditions @exeterage.bsky.social Follow us here and see our GitHub for resources and updates bsky.app/profile/lcpi...

000

Luke Pilling @lcpilling.bsky.social · 08/05/2025

Overview of our genetics of multimorbidity (GEMINI) project by @timfrayling.bsky.social at the MLTC symposium We are studying 72 common long-term chronic conditions with a heritable (genetic) component. Our seminal paper is now published (doi.org/10.1016/j.eb...) and our public resources are below:

Professor Tim Frayling presenting the GEMINI slides

311

Luke Pilling @lcpilling.bsky.social · 08/05/2025

Having a great time at a multimorbidity symposium in Birmingham this week. Many interesting talks on care pathways, drug targets, etc Pictured is the amazing @janemasoli.bsky.social talking about the important work in the clinical community of practice @exeterage.bsky.social @exeter.ac.uk

020

Luke Pilling @lcpilling.bsky.social · 21/01/2025

Quick way to get extreme p-values in R is to return them as strings! log_pvalue = log(2) + pnorm(abs(z),lower.tail=FALSE,log.p=TRUE) log10_pvalue = log_pvalue/log(10) ## from LN to log10 mantissa = 10^(log10_pvalue %% 1) exponent = log10_pvalue %/% 1 sprintf("%1.2fe%d",mantissa,exponent)

Screenshot of my R console running the few commands to get character/string output of extreme p-values

251

Luke Pilling @lcpilling.bsky.social · 09/01/2025

My blood 🩸 made it all the way to University Hospital, Birmingham where it helped to save or improve someone's life Love getting these e-mails. Consider signing up to donate blood, if you don't already www.blood.co.uk @givebloodnhs.bsky.social

Map showing the route from Exeter up to the centre in Filton where they ran tests and then processed it into a form that can be given to a patient in need. Once it was ready it was sent to the University Hospital, Birmingham where it will help to save or improve someone's life.

061

Luke Pilling @lcpilling.bsky.social · 10/12/2024

Daniel Williamson at the UK Biobank scientific conference explaining his study of MRI-estimed cortical brain iron and incident dementia risk @ukbiobank.bsky.social #UKBSC24

040

Luke Pilling @lcpilling.bsky.social · 09/12/2024

So much anticipation for this session at the UK Biobank scientific conference tomorrow #UKBSC24 If any of you are going, reach out: let's meet and chat! @ukbiobank.bsky.social #AcademicSky

183

Luke Pilling @lcpilling.bsky.social · 15/11/2024

Really enjoying the biological science sessions at #GSA2024 - room 4C-3 is where it is at! Current fascinating talk is by Patricia Opresko on how oxidative damage to #telomeres promotes #senescence

030

Luke Pilling @lcpilling.bsky.social · 07/10/2024

Latest version of {ukbrapR} is mostly to improve documentation (plus 2 bug fixes) Website now includes "dark mode" and "Articles" for common uses analyzing UK Biobank on DNAnexus using #Rstats! lcpilling.github.io/ukbrapR/

Screenshot of the ukbrapR package GitHub website, showing "dark mode" toggled, and the Articles drop-down menu

030

Luke Pilling @lcpilling.bsky.social · 09/01/2024

Great start to 2024: paper published! tiny.cc/x0eyvz Iron and risk of dementia: Mendelian Randomisation analysis in UK Biobank Genetically predicted higher serum iron increased risk of non-AD dementia in European and African ancestry groups, supporting a causal link. This was independent of HFE

Screenshot of the paper abstract. We used Transferrin Saturation as the clinical marker of iron

Table 2 from our new paper, showing that the genetically predicted transferrin saturation associations with dementia is driven by clinically defined non-Alzheimer's diagnoses. Lots of caveats regarding the use of polygenic scores in non-EUR groups, but we aimed to apply the principles outlined in the recent report to the best of our current abilities, but future studies should move away from discrete ancestry clusters https://doi.org/10.17226/26902

120

Luke Pilling @lcpilling.bsky.social · 10/11/2023

Been waiting 15 minutes already. Somethings tells me this is not going to be a productive morning 😓 #BlueScreenOfDeath

000

Luke Pilling @lcpilling.bsky.social · 26/10/2023

Once in a JupyterLab environment on a Spark Cluster the following R commands pull the data straight from the database without intermediate files Takes ~5 minutes to build all the package dependencies, and <5 minutes to get the data object

000