Sign in

Christian Bartels

@chriba.bsky.social
126 followers 221 following 61 posts

Interested in data ... pharmacometrics, causal inference, ..., risk management, cheminformatics, molecular modeling, protein structure determination, peptide sequencing | Basel scholar.google.com/citations?user=R…

PostsRepliesMedia
Reposted by Christian Bartels
Eric Topol @erictopol.bsky.social · 05/09/2026
Why did 2 large, pivotal clinical trials [vs interleukin-6 and Lp(a)] fail to lower heart attacks, strokes and cardiovascular deaths? erictopol.substack.com/p/the-corona...
erictopol.substack.com
The Coronary Artery Inflammation Controversy
How Two Large Outcome Trials Based on Surrogate Metrics Failed
09119
Reposted by Christian Bartels
Joe Noonan @joenoonan.bsky.social · 07/08/2026
Uruguay, with 3.5 million people, generates 98% of its electricity from renewable sources. The architect of that transformation, Ramón Méndez Galain, put it simply: "The key is not technology; it is institutions. Once the rules are fair and predictable, the system builds itself."
irishexaminer.com
Our solar power model isn't working — there's a fairer way
Solar lease-to-own could unlock Ireland's energy future — Uruguay, the Netherlands and Germany have all made this work with straightforward rules, writes Mary Teehan
3331421073
Reposted by Christian Bartels
Jan Drugowitsch @jdrugowitsch.bsky.social · 13/07/2026
🚨 Big news 🚨 After 10 years at Harvard, my lab is moving to Barcelona in Spring 2027 to join @cbc-upf.bsky.social and the Department of Engineering at @upf.edu. 1/4
68818
Christian Bartels @chriba.bsky.social · 19/06/2026
Causal inference without do-calculus nor potential outcomes. An important simplification? #causal #causal_inference #swig #docalculus #identification
020
Reposted by Christian Bartels
Hadley Wickham @hadley.nz · 19/06/2026
Coding agents like Claude Code feel like magic but they’re really a bunch of tools in a trenchcoat. This week I build a working coding agent in R, starting from just three tools. I’ll add some extra tools to make it safer and more efficient. tidydesign.substack.com/p/a-coding-a...
tidydesign.substack.com
A coding agent is six functions in a trenchcoat
Coding agents like Claude Code, Cursor, and Codex have taken the software engineering field by storm. What makes them tick?
79616
Reposted by Christian Bartels
Our World in Data @ourworldindata.org · 19/06/2026
Many of us can save a child's life, if we rely on the best data. Giving money to a charity is one of the best things you can do for others. Your donation can make a huge difference. But whether or not your donation makes a difference greatly depends on *where* you donate.
Bar chart of perceived and expert estimates of charity cost-effectiveness where laypeople think the most effective charity is about 1.5 times as effective as the average charity, while global health experts estimate about 100 times as effective. The chart highlights a large gap between public perception and expert judgment. Data source: Caviola et al. (2020), OurWorldinData.org. License: CC BY to Max Roser.
2446
Reposted by Christian Bartels
Thilo Muth @drmuth.bsky.social · 15/06/2026
New preprint on the current bottlenecks of deep learning-based methods for de novo peptide sequencing that underuse physical properties of spectra and how to overcome this with MemNovo: arxiv.org/abs/2606.11868
arxiv.org
MemNovo: Look Back at the Spectrum for Balanced De Novo Peptide Sequencing from Mass Spectrometry
De novo peptide sequencing from tandem mass spectrometry is pivotal in proteomics, enabling identification of novel peptides without reference databases. While recent Transformer-based encoder-decoder...
053
Reposted by Christian Bartels
Jess Nevins @jessnevins.bsky.social · 14/06/2026
These three players from the Ivory Coast's World Cup team walked off the airplane looking like this. I'd say that's grossly unfair to the rest of humanity. (Ivorian designer Ibrahim Fernandez).
Three *extremely* well-dressed futbol players from Ivory Coast.
1927456700
Reposted by Christian Bartels
Fabrice aka Curious Crocodile @curiouscrocodile.bsky.social · 14/06/2026
Défaite nette de l’UDC en Suisse, l’initiative « Pas de Suisse à 10 millions » est heureusement rejetée à 55%. Le parti d’extrême-droite annonce bien sûr déjà qu’il n’en restera pas là et gagnera la prochaine fois: la démocratie c’est seulement quand ça les arrange. Continuons à les faire perdre!!
15313
Reposted by Christian Bartels
The Economist @economist.com · 14/06/2026
A population cap would disrupt Switzerland’s relationship with Europe. Register for free to read how the country should instead regulate the inflow of migrants without breaking EU agreements
econ.st
The Swiss would be foolish to cap their population at 10m
Immigration sometimes needs brakes, but the proposal would be like driving into a wall
264
Christian Bartels @chriba.bsky.social · 12/06/2026
Early us of Bayesian approach in bioinformatics. Use available knowledge on chemical shifts and distances in homologous proteins as prior to assign NMR spectra for protein structure determination. en.wikipedia.org/wiki/Structu... doi.org/10.1007%2FBF... #nmr #protein #structure #bayesian
011
Christian Bartels @chriba.bsky.social · 07/06/2026
Early use of directed acyclic graphs to speed up analysis of mass spectrometry of peptides. Low polynomial time complexity, down from exponential complexity without DAGs. en.wikipedia.org/wiki/De_novo... doi.org/10.1002/bms.... #DAG #Peptide #sequencing
231
Reposted by Christian Bartels
Theiss Bendixen @theissbendixen.bsky.social · 06/06/2026
It's alive! 🎉 𝗧𝗵𝗲 𝗗𝗮𝘁𝗮 𝗔𝗻𝗮𝗹𝘆𝘀𝘁'𝘀 𝗚𝘂𝗶𝗱𝗲 𝘁𝗼 𝗖𝗮𝘂𝘀𝗲 𝗮𝗻𝗱 𝗘𝗳𝗳𝗲𝗰𝘁 is out -- an introduction to causal inference in practice. The first two chapters are available for free here: theissbendixen.com/dag-book/ More below 👇
38522
Reposted by Christian Bartels
Eva Herzog @evaherzog.bsky.social · 07/06/2026
In unsicheren Zeiten brauchen wir stabile Beziehungen zu Europa. Aber die Chaos-Initiative reisst genau diese Brücken ein – sie ist politisch und wirtschaftlich ein Eigentor, das unsere Sicherheit und unseren Wohlstand gefährdet.
1126
Christian Bartels @chriba.bsky.social · 02/06/2026
Single-world intervention graphs (SWIGs) as distributions. Systematic way to derive identifying expressions for estimands. New front-door derivations extending more readily to complex settings. doi.org/10.48550/arX... #SWIG #DAG #causal #identification #frontdoor
doi.org
Single World Intervention Graphs as Distributions: A Framework for Causal Identification
Causal inference seeks to estimate the effect of an intervention on an outcome using observed data, typically via Rubin's potential-outcome framework or Pearl's do-calculus. Following section 9 of Ric...
100
Reposted by Christian Bartels
Michael "Shapes Dude" Betancourt @betanalpha.bsky.social · 24/05/2026
Everything is a modeling problem.
A probabilistic graphical model of a data generating process before an intervention.A probabilistic graphical model of a data generating process after an intervention.A probabilistic graphical model of both data generating processes jointly.The predictive distribution for observations after the intervention given data from before the intervention derived from the joint model.
1183
Reposted by Christian Bartels
Oliver Maclaren @omaclaren.bsky.social · 05/05/2026
Finally uploaded a revised version of this preprint on parameter identifiability arxiv.org/abs/2502.04867. Approach called 'invariant image reparameterisation' though 'image' used in mathematical sense not picture sense. Took me a while to revise for various reasons, but fairly happy with it now
1114
Reposted by Christian Bartels
Andrew Heiss @andrew.heiss.phd · 22/05/2026
Finally switched to DuckDuckGo and it’s nice to not have LLM summaries anymore
Screenshot from Arc’s preferences showing DuckDuckGo as the default search engine
1152060186
Reposted by Christian Bartels
Our World in Data @ourworldindata.org · 18/05/2026
How will populations across the world change in the 21st century? 🔧 Explore for yourself with our new interactive tool!
An image of the new interactive population simulation tool from Our World in Data, showing South Korea.
410233
Reposted by Christian Bartels
CAUSALab @causalab.org · 31/03/2026
One week away! Don't miss the next Methods Series presentation with Alex Ocampo.
022
Reposted by Christian Bartels
Our World in Data @ourworldindata.org · 14/03/2026
The median age in China has rapidly caught up with the United Kingdom— In 1965, the median age in the United Kingdom was almost twice that of China. Half of the people in the UK were younger than 34 years, and half were older. In China, this midpoint was just 18 years.
The median age in China has rapidly caught up with the United Kingdom.

Line chart of median age for China and the United Kingdom from 1950 to 2025, with the vertical axis in years from 0 to 40 and the horizontal axis showing years 1950 to 2025. A line labeled United Kingdom stays around mid-30s in 1950, dips slightly to about 33 by the mid-1970s, then gradually rises to about 40 by 2025. A line labeled China starts around 22 in 1950, falls to about 18 to 19 in the mid-1960s and 1970s, then climbs steadily to meet the UK at about 40 in 2025. Annotated note: in the mid-1960s China’s median age was just under half that of the UK; another note states that today the median age in both countries is 40 years. Data source: UN, World Population Prospects (2024). License: CC BY.
26713
Reposted by Christian Bartels
axaV @axaxaxav.eurosky.social · 27/05/2024
least sexist industriebranche
nzz-artikel: "IT-Branche Zürich lockt Frauen: Pink
Die IT-Branche will mehr Frauen für sich gewinnen – mit viel Pink und Rosa und «<einer bildlichen, einfachen Sprache>>
Eine Studie, unterstützt vom Kanton Zürich, <deckt die wahren Empfindungen von Mädchen und Frauen auf».
Zeno Geisseler
26.05.2024, 05.35 Uhr 3 min"

darunter ein bildausschnitt einer mutmaßlichen frau mit pinkem hemd (man sieht nur einen teil des torsos, arme und hände), die in einem büro sitzt und einen laptop bedient
1112319
Reposted by Christian Bartels
Michael "Shapes Dude" Betancourt @betanalpha.bsky.social · 07/03/2026
The Geordi LaForge meme demonstrating hesitation for a conventional “causal” direct graphical model, showing only observed variables and unobserved confounders, and enthusiasm for a directed graphical model that represents a full joint probabilistic model including all observed variables and all latent/unobserved variables.
0165
Reposted by Christian Bartels
Jeremy Cliffe @jeremycliffe.bsky.social · 06/03/2026
Meanwhile, Spain's massive investment in renewables is paying dividends now: with prices for Spanish industry and consumers low and stable compared with other European economies. www.ft.com/content/ac77...
281746763
Christian Bartels @chriba.bsky.social · 28/02/2026
Post a pic you took, no context, to bring some zen to the feed.
110
Reposted by Christian Bartels
Andrew Heiss @andrew.heiss.phd · 23/02/2026
spending my sunday evening once again attempting to draw a DAG for diff-in-diff
DAG representing the causal structure of a standard difference-in-differences design with two locations and two time periods—units in one location in the post-period receive treatment. $L$ = group or location indicator (treated vs. untreated location); $T$ = time indicator (pre vs. post period); $U$ = unobserved time-invariant confounders (e.g., GDP per capita, general health status, public health infrastructure). $X \leftarrow T \rightarrow Y$ represents a common time trend affecting both locations equally. The causal effect of $X$ on $Y$ is identified by conditioning on $\{L, T\}$, which corresponds to using location and time indicator variables in a regression like `y ~ location * period`.DAG representing the causal structure of a standard difference-in-differences design, but with explicit pre- and post-treatment outcomes. $L$ = group or location indicator (treated vs. untreated location); $T_\text{post}$ = post-period measurement (indicator that the observation occurs after the intervention); $X_\text{post}$ = treatment (which only occurs for treated locations in the post period); $Y_\text{pre}$ and $Y_\text{post}$ = outcome measured before and after the intervention. $U$ = unobserved time-invariant confounders (e.g., GDP per capita, general health status, public health infrastructure). $Y_\text{pre} \rightarrow Y_\text{post}$ represents outcome persistence (e.g. autocorrelation or slow-moving changes); $X_\text{post} \leftarrow T_\text{post} \rightarrow Y_\text{post}$ represents a common time trend affecting both locations equally. The causal effect of $X_\text{post}$ on $Y_\text{post}$ is identified by conditioning on $\{L, T_\text{post}\}$, which corresponds to using location and time indicator variables in a regression like `y ~ location * period`.
108411
Reposted by Christian Bartels
Martin Huber @causalhuber.bsky.social · 23/02/2026
📣The 2026 Symposium of #CausalInference in the #HealthSciences takes place on March 18, 2026 in Fribourg. Theme: AI & machine learning in causal inference for health sciences 🎤Keynotes: Elsa Gautrain, Aurélien Sallin, Jonas Peters, Jana Mareckova 🔗https://projects.unifr.ch/pophealthlab/?page_id=1561
041
Christian Bartels @chriba.bsky.social · 22/02/2026
We apply the estimand framework to dose–exposure–response analyses. ... strategy to improve exposure–response analyses for dose selection, particularly when the relevant evidence includes data from multiple studies. #estimand #exposure-response #dose-response #causal doi.org/10.1002/psp4...
doi.org
Some Common Dose–Exposure–Response Estimands and Conditions for Their Causal Identifiability
Exposure–response analyses are central to dose selection in drug development. The estimand framework, formalized in ICH E9(R1) regulatory guidance, provides a structured approach to define scientific...
000
Reposted by Christian Bartels
Pausal Zivference @pausalz.bsky.social · 09/10/2025
For years I had trouble following some of the discussion about confidence bands, but at ACIC this year @noahgreifer.bsky.social pointed me to a helpful paper So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas arxiv.org/abs/2510.07076
arxiv.org
Confidence Regions for Multiple Outcomes, Effect Modifiers, and Other Multiple Comparisons
In epidemiology, some have argued that multiple comparison corrections are not necessary as there is rarely interest in the universal null hypothesis. From a parameter estimation perspective, epidemio...
1208
Reposted by Christian Bartels
Noah Greifer @noahgreifer.bsky.social · 18/02/2026
I'm so excited to announce the first release of my newest #Rstats package, {adrftools}! This package facilitates estimation, visualization, and testing for the causal effect of a continuous (i.e., non-discrete) treatment. 🧵 1/10 #statssky #episky #causalinference
cran.r-project.org
adrftools: Estimating, Visualizing, and Testing Average Dose-Response Functions
Facilitates estimating, visualizing, and testing average dose-response functions (ADRFs) for characterizing the causal effect of a continuous (i.e., non-discrete) treatment or exposure. Includes suppo...
511530
Reposted by Christian Bartels
Daniel Lakens @lakens.bsky.social · 17/02/2026
As I said, this is the important work that needs to be done. Mark's paper is overly simplistic, arguing we can't judge deviations at all. Deviations are in practice not remotely as bad as he wants. If he had collected actual data,che would have falsified his own claims.
101
Reposted by Christian Bartels
Paul Hünermund @p-hunermund.com · 08/07/2025
This is a fairly technical but highly relevant paper on how we can model complex systems at various levels of detail without losing causal content. Think gas: instead of tracking every molecule, we can focus on big-picture properties like temperature and pressure. www.auai.org/uai2017/proc...
1016935
Reposted by Christian Bartels
Dr Will Ball @wpball.com · 30/01/2026
What are some of the best DAGs you seen that depict time-varying confounding? Could be from a 'real' worked example, or generic. #EpiSky
6125
Reposted by Christian Bartels
ArXiv Paperboy (Stat.ME+Econ.EM) @paperposterbot.bsky.social · 13/01/2026
arXiv📈🤖 Long-Term Causal Inference with Many Noisy Proxies By Lal, Imbens, Hull
We propose a method for estimating long-term treatment effects with many short-term proxy outcomes: a central challenge when experimenting on digital platforms. We formalize this challenge as a latent variable problem where observed proxies are noisy measures of a low-dimensional set of unobserved surrogates that mediate treatment effects. Through theoretical analysis and simulations, we demonstrate that regularized regression methods substantially outperform naive proxy selection. We show in particular that the bias of Ridge regression decreases as more proxies are added, with closed-form expressions for the bias-variance tradeoff. We illustrate our method with an empirical application to the California GAIN experiment.
0115
Reposted by Christian Bartels
Justin @ediparolev.bsky.social · 10/01/2026
“Uruguay did what most nations still call impossible: it built a power grid that runs almost entirely on renewables—at half the cost of fossil fuels. The physicist who led that transformation says the same playbook could work anywhere—if governments have the courage to change the rules.”
forbes.com
Uruguay’s Renewable Charge: A Small Nation, A Big Lesson For The World
Uruguay built a power grid that runs 99% on renewables—at half the cost of fossil fuels. Here’s how its bold energy overhaul became a global model.
208114544677
Reposted by Christian Bartels
Altmetric @altmetric.com · 08/01/2026
X/Twitter's rough full volume is around 500 total million posts every day, or 182 (and a half) billion posts per year. By contracts, we found 11.2 million research posts in all of 2025 on there. In other words, 0.000006% of Twitter appears to be sharing research. Basically zero.
415647
Christian Bartels @chriba.bsky.social · 04/01/2026
This gives financial institutions leeway to acquire hard-to-sell, higher-yielding long-term assets and finance them with cheaper short-term liabilities, thus increasing profits through a risky “mismatched book”.
000
Christian Bartels @chriba.bsky.social · 02/01/2026
000
Reposted by Christian Bartels
ArXiv Paperboy (Stat.ME+Econ.EM) @paperposterbot.bsky.social · 01/01/2026
link 📈🤖 Demystifying Proximal Causal Inference (Ringlein, Nguyen, Zandi et al) Proximal causal inference (PCI) has emerged as a promising framework for identifying and estimating causal effects in the presence of unobserved confounders. While many traditional causal inference methods rely on the
032
Christian Bartels @chriba.bsky.social · 31/12/2025
Happy new year! As an idea for improved privacy pr.tn/ref/X11M0Y8Q
pr.tn
Enjoy two weeks of Proton for free
Experience true online privacy. Protect your data with Proton for free for the next 14 days.
000
Reposted by Christian Bartels
Lumo @asklumo.proton.me · 26/11/2025
Lumo now understands LaTeX. Type any formula (fractions, matrices, integrals, etc.) straight into chat and see it rendered instantly. No more screenshots or ASCII tricks. Perfect for precise technical work and academic tasks. Try it now: lumo.proton.me
Lumo rendering an integral in LaTeX.
1463
Reposted by Christian Bartels
Moritz Schauer @mschauer.bsky.social · 19/12/2025
Students learn early that correlation does not imply causation. Correct, but incomplete... Certain patterns of independence and conditional dependence constrain #causal structure very strongly. A simple example:
142
Reposted by Christian Bartels
Cliff PervoSCARY @pervocracy.bsky.social · 15/12/2025
The older I get, the more my politics mature from childish, naïve beliefs like "the world is complicated and leaders have to make hard decisions" to more serious, adult principles like "hurting people is bad and helping people is good."
45253485500
Reposted by Christian Bartels
Reiner @erlangen.bsky.social · 16/12/2025
IL-17-mediated antifungal immunity restricts Candida albicans pathogenicity in the oral cavity www.nature.com/articles/s41... The microbiome not only consists of bacteria, but also of fungi. Most of them support human and animal health medicalxpress.com/news/2025-12... However, some fungi ...
nature.com
IL-17-mediated antifungal immunity restricts Candida albicans pathogenicity in the oral cavity - Nature Microbiology
IL-17 signalling restricts C. albicans pathogenicity in the colonized oral cavity. Lack of IL-17 is associated with overt filamentation due to impaired zinc nutritional immunity and over time leads to...
063
Reposted by Christian Bartels
The Economist @economist.com · 02/12/2025
Despite claiming to have seized full control of Pokrovsk, it is not clear that Russia has done so—or that it can necessarily press on for longer than Ukraine can hold on. Our charts break down the data
econ.st
Ahead of peace talks, Russia’s battlefield advances remain slow
Even at an accelerated recent pace, seizing Ukraine’s eastern regions would take more than two years
1105
Christian Bartels @chriba.bsky.social · 23/11/2025
#datascience: support decisions based on available data #statistics: discuss data that is required to make reliable decisons ... and hands off ...
010
Reposted by Christian Bartels
The Economist @economist.com · 22/11/2025
The 28-point proposal being hawked around by America is so poorly put together, so vague, unbalanced and impractical that, in a more normal world, it would never have seen the light of day
econ.st
Donald Trump’s peace plan would be bad for Ukraine, Europe and America
It is a sad mix of naked opportunism and strategic myopia
2922096
Reposted by Christian Bartels
PLOS Medicine @plosmedicine.org · 20/11/2025
We are thrilled to now be on BlueSky! We prioritize studies addressing critical health challenges—from major diseases to health equity—that bridge scientific rigor with real-world impact, connecting researchers, clinicians, and policymakers to advance global health 🧬🧪 #OpenScience #MedSky
39838
Reposted by Christian Bartels
Howard French @hofrench.bsky.social · 13/11/2025
More than 97 percent effective. www.npr.org/sections/goa...
npr.org
New malaria drug could be a life-saver as the standard drug shows signs of weakness
The best drug to fight malaria is facing increased resistance from the parasites it fights. Now there's an alternative in the pipeline and it looks promising.
1369
Reposted by Christian Bartels
Juha Karvanen @juhakarvanen.bsky.social · 13/11/2025
In longitudinal studies, dropout leads to a monotone missing data pattern. We show in a new article that monotonicity sometimes enables and sometimes prevents the identification of  the full law, i.e., the joint distribution of actual variables and response indicators. openreview.net/pdf?id=kVthd...
Two graphs where monotonicity prevents identification
063