@slowkow.com
Computational biologist 🧬🖥️ at Mass General Brigham and Broad Institute, PhD at Harvard. Bioinformatics, transcriptomics, COVID, genomics, immunology, genetics, statistics, and web development. I made #ggrepel and I blog at https://slowkow.com
The last part abt how measles disrupts immunity vs other infections is, imo, underappreciated (& mechanism was recently clarified). It's part of why measles vaccination was seen to help beyond just reducing measles complications - not getting measles reduces severity of subsequent infections too.
Overjoyed to share our new work exploring the antigen specificity of CSF-expanded CD8+ T cells in #multiplesclerosis #EBV in @natimmunol.nature.com #immunology 🧪🧵1/
www.nature.com/articles/s41...
Diagram of melanoma cell immunotherapy, detailing mechanisms of action of drugs like Ipilimumab, Nivolumab, and Vemurafenib. Highlights interactions between immune cells (T cells) and melanoma cells, targeting pathways like CTLA-4, PD-1, and MAPK signaling.
The incidence and prevalence of #melanoma, the fifth most common cancer in the US, have increased over the last 5 decades.
This Review summarizes the epidemiology, pathophysiology, diagnosis, and treatment of cutaneous melanoma.
ja.ma/4j1AhEC
We are very excited to share the first preprint of a new direction for our group. Led by the fearless duo of @arthurwchow.bsky.social and @hoyinchu.bsky.social, our foray into computational protein design—
This is shocking stuff. John Quackenbush has been brilliant since forever. www.statnews.com/2025/12/05/r...
You run a new command in #rlang #python #bash, and you don’t really know how long it should take.
Will it be done in 30 seconds? 5 minutes? 45 minutes? Longer? 😫
An automatic notification might help to stay focused on #programming
Let me introduce you to ntfy.sh
slowkow.com/notes/ntfy/
Cruz Godar shares stunning Javascript and webGL applets that demonstrate the beauty of mathematics.
cruzgodar.com
A bad thing is unfolding at NIH this week: It looks like the Trump administration is trying to replace key civil servant scientific leaders, the Institute Directors, with political hires. These directors control the NIH budget, tens of billions.
A bit of a video explainer here: 1/ 🧪
Check out our new paper @nejm.org. RSV, COVID-19 & Influenza vaccines continue to show consistent effectiveness & safety based on a review & synthesis of 500+ research studies👇! We worked hard to evaluate the evidence so you can be confident about getting vaccinated this fall!
South Carolina schools recently sent home unvaccinated kids to stem a growing measles outbreak. It turned out 5 of them were infected and, if they'd been at school, could've unwittingly spread it to their classmates for days.
That kind of precaution is now illegal in Idaho.
Your work is very inspiring, thank you for sharing!!
New release of PowerLMM.js! Browser-based power analysis for longitudinal models with dropout.
Now includes:
- Power analysis summary report
- Reproducible & shareable configs (URL/JSON)
- Calculations validated against R
- Hypothesis region visualization
powerlmmjs.rpsychologist.com
Long Covid Is Real — And It’s Changing an Entire Generation
Hundreds of thousands of kids in America are struggling with an illness that many doctors and schools refuse to recognize.e
Feature: www.rollingstone.com/culture/cult...
A visualization of average resting heart rate in Germany from 2020 through 2022.
Link: corona-datenspende.github.io/en/vitaldata...
Data: zenodo.org/records/8229...
A huge thank you to my coauthor Pritha Sen, mentor Chloe Villani, colleague Chris Cosgriff, and the whole COVID-19 Severity team for making this study possible and supporting me all these years. I am very grateful to the hundreds of people who created this opportunity to study disease severity.
**M.** Volcano plot for convalescent CD4 T cell subset 6 (Treg). Dots indicate genes, x-axis indicates fold-change, y-axis indicates negative log10 p-value from multivariate linear regression. Red color indicates FDR < 5%. **N.** Box plot of *ICOS* gene expression in CD4 T cell subset 6 (Treg) in the convalescent data. Fold-change and p-value is shown. Dots represent patients, box plots indicate median and interquartile range, x-axis and hue represent convalescent COVID infection. **O.** Box plot of *ICOS* gene expression in CD4 T cell subset 13 (Treg) in the acute data. Dots represent patients, box plots indicate median and interquartile range, x-axis represents time point, and hue represents COVID infection.
While many immunologic abnormalities in acute severe COVID-19 resolve during convalescence 3-months post-infection, we observed persistently high ICOS expression in regulatory T cells, potentially linking acute infection to chronic post-COVID syndromes.
**E.** Amino acid position associations with COVID severity for HLA-DQB1. x-axis indicates position along the gene. y-axis indicates negative log10 p-value from multivariate linear regression. Black color indicates FDR < 5%. **F.** Coefficients from the multivariate linear regression for the effect size of each amino acid position on COVID severity. Amino acid positions D57 and V57 are highlighted. Error bars indicate 95% CI. **G.** Bar plots of the number of patients with each genotype. Facets indicate genotype (0, 1, 2) for amino acid position D57 (top) and V57 (bottom). x-axis indicates number of patients. Hue indicates COVID severity. **H.** Box plot of abundance of CD8 T cells with TRBV28, x-axis indicates time point and hue indicates genotype of HLA-DQB1 V57 (0, 1, 2). P-value from multivariate linear regression.
We also identify HLA-DQB1 amino acid positions associated with:
- COVID-19 disease severity
- specific TCRs in CD8 T cells
- viral load
- neutralization capacity of serum
A scatter plot where each point represents a gene. The x-axis represents the fold-change of the gene in acute patients who received tocilizumab treatment, where genes induced after treatment are seen on the right and genes repressed after treatment are seen on the left. The y-axis represents the fold-change of the gene in acute patients, where genes associated with worse severity are seen on the top, and genes associated with lesser severity are seeon on the bottom. Some genes are labeled.
We found that tocilizumab eliminates CLU-expressing MDSCs and ISG-positive myeloid subsets, restores antigen presentation, and reactivates productive adaptive immunity.
In myeloid cells, the mRNA signature of tocilizumab treatment appears inverse to the signature of disease severity.
A schematic that shows some factors associated with the severity of COVID-19. Starting with SARS-CoV-2 infection, the adaptive immune response should lead to successful elimination and convalescence. In some patients, various factors such as viral titer, age, autoantibodies, lymphopenia, serum IL-6 levels, TCR and BCR clones, and HLA genotypes can increase risk for severe disease. This leads to a feedback loop wherein lack of viral clearance leads to increased tissue damage. This is associated with myeloid cell dysfunction marked by impaired antigen presentation, which drives a non-productive adaptive immune response, as reflected by reduced expression of B and T cell gene programs involved in antigen recognition, immune synapse formation, and cytotoxicity. The anti-IL-6R antibody, tocilizumab, appears to reverse the mRNA signature of disease severity in myeloid cells.
Severe disease is also linked to autoantibodies targeting type I interferons, influenced by specific HLA-DQB1 allelic variants, and strongly correlated with serum IL-6 levels.
Here is a schematic representation of our findings:
Our findings show that myeloid dysfunction, which is marked by impaired antigen presentation, drives a non-productive adaptive immune response, as reflected by reduced expression of B and T cell gene programs involved in antigen recognition, immune synapse formation, and cytotoxicity.
We analyzed 2.5 million immune cells from 428 patients in three contemporaneous SARS-CoV-2 cohorts:
(1) acutely infected patients across severity levels and time points
(2) patients from a trial to study tocilizumab for COVID-19
(3) convalescent patients three months after infection
Overview of a research study about COVID-19. The study includes three cohorts: (1) 351 acute COVID-19 patients across 5 severity levels and 3 time points profiled by multiple modalities, (2) 4 acute COVID-19 patients treated with tocilizumab, (3) 73 patients, 43 of whom were previously infected with SARS-CoV-2 3 months ago. The overview shows the general study design and visualizations of the single-cell RNA sequencing data analyzed in this study.
I'd like to announce the medRxiv preprint of our latest work:
A multimodal atlas of COVID-19 severity identifies hallmarks of dysregulated immunity.
www.medrxiv.org/content/10.1...
Search #Pubmed and view the results on a 2D embedding. Each point represents a paper, and similar papers are near each other. Colors indicate clusters of similar papers.
By Viktor Petukhov
github.com/multicore-ca...
AI slop and the destruction of knowledge irisvanrooijcogsci.com/2025/08/12/a...
1. Make an Excel file with your author information
2. Drag and drop
3. Copy a nicely formatted author list into Word
🐶 slowkow.com/authorbud
A graph from a 1993 article published in the New England Journal of Medicine illustrating the relationship between air pollution and mortality rates in six U.S. cities. The x-axis represents the concentration of fine particles (µg/m³), while the y-axis shows the adjusted mortality-rate ratios. Cities such as Steubenville, OH, and Portage, WI, are marked. The adjusted mortality rate for the most polluted of the cities (Steubenville, OH) is 1.26 times higher than the morality rate least polluted city (Portage, WI). Data from Dockery et al., NEJM 1993.
"After adjusting for smoking and other risk factors, we observed statistically significant and robust associations between air pollution and mortality."
"Air pollution was positively associated with death from lung cancer and cardiopulmonary disease"
pubmed.ncbi.nlm.nih.gov/8179653/
Convert Markdown to C++, Python, MATLAB, and LaTeX
iheartla.github.io
by Yong Li
Could you please share the original links or files for the source data you are using in these figures?
The summer fight for science. The time is now. Join the fight at standupforscience.net.
☀️THE SUMMER FIGHT FOR SCIENCE☀️
🫵 WHO: You + all our friends
🧪 WHAT: Share science with your neighbors and tell them about the impact of proposed budget cuts.
🇺🇸 WHERE: Your local community
🏖️ WHEN: Now to Sept 30th
🔬 WHY: The 9/30 budget vote will determine the future of science in America.
