"Biology is…about propensities, potentials, vulnerabilities, predispositions, proclivities, interactions, modulations, contingencies, if/then clauses, context dependencies, exacerbation or diminution of preexisting tendencies.”
- Robert Sapolsky, from his book Behave
What a f-in great quote!
🚨 New pre-print! 🚨 In the largest study of its kind to-date, we investigate the ecological and evolutionary mechanisms driving within-patient evolution of antimicrobial resistance (AMR). Read here:
www.biorxiv.org/content/10.6... , and follow along with this thread, discussing our findings (1/21)
Here is an announcement for everybody who is sick and tired on the attacks on science and scientists.
It has been months in the making, but now we are ready to fight back.
First episode airs tomorrow 5pm UK.
New episode every friday.
Incredible guests lined up.
open.substack.com/pub/protagon...
Lost in translation
It's actually snowing in Amsterdam today ha
amsterdamexperiences.nl/blog/ai-and-...
12:37 and still going strong…sounds like 112 is blitzed
Learned today that New Year’s Eve in Amsterdam is absolutely insane
Cornwall, Oxford and London Holidays!
OpenEvidence valued at >$10B with $100M in revenue. Not much in this post about if it improves patient care..but that's old hat
www.linkedin.com/posts/ethan-...
No just EHR data. Good idea for the next release
This month's @natrevmicro.nature.com is fantastic!
Here @amsterdamumc.bsky.social, my team is building from scratch a next-gen clinical micro informatics infrastructure to support real-time pathogen surveillance; explicitly connecting the hospital to public health
www.nature.com/nrmicro/volu...
Links to the other 3 datasets:
Stanford:
datadryad.org/dataset/doi:...
University of Texas Southwestern:
datadryad.org/dataset/doi:...
Eastern Caroline University Health:
datadryad.org/dataset/doi:...
(6/6)
Thank you to Jonathan Chen at Stanford, Rich Medford at UTSW / ECU, @alexwei21.bsky.social, Physionet, Mass General Brigham, NIH and all other team members who worked on this monumental, multiyear effort! (5/6)
I hope this release will inspire other researchers working with observational data (in any area), to go through the trouble of releasing de-identified patient-level data. Science is beset by issues of reproducibility, transparency & trust. This approach is one clear way to help right the ship. (4/6)
The dataset is hosted by Physionet (where MIMIC lives). Best of all, ARMD-MGB is harmonized with 3 other datasets *of equal size and detail* from Stanford, UTSW, and Eastern Carolina U, making this by far the largest, most detailed, and geographically diverse dataset ever for the study of AMR. (3/6)
The ARMD-MGB is a de-identified dataset engineered specifically for the study of AMR. It includes highly granular microbiology data (MICs / KBs, testing methods, etc) for over 970K samples, and extensive clinical metadata (prior meds, comorbidities, etc) from >226K patients. (2/6)
Today I am very proud to announce the release of the Antibiotic Resistance Microbiology Dataset - Mass General Brigham (ARMD-MGB; physionet.org/content/armd...) , as part of an NIH-funded collaboration led by Jonathan Chen at Stanford. (1/6)
Sign me up!!
When you can watch an entire series with a thoroughly unlikeable main character, you know the director and actor have done an amazing job!
Also, the attention to detail in the set pieces somehow manages to capture the beauty of that time and place without empty sentimentality, brilliant
It is hard to know what durations of treatment best prevent the selection of resistance at the population level; this is partly on the basis of data that suggests most people develop resistance post treatment through exogenous colonization of a damaged microbiome, rather than endogenous evolution.
www.huffpost.com/entry/histor...
No easy solutions, but a lot of excellent points. It may be that tomorrow’s leaders will be those who have the courage to fail
Now out in @natcomms.nature.com Kudos to @tylim.bsky.social and @jameshay.bsky.social for a huge effort and thanks to all the collaborators for their hard work. See the final version here: www.nature.com/articles/s41...
This week is World #AntimicrobialResistance Awareness week
This article collection spans the breadth of #AMR across pathogens (bacteria, viruses, fungi, parasites), covers epidemiological monitoring, diagnostics, treatment, etc
Lancet infographic summarising the key findings and recommendations from The Lancet Infectious Diseases Series on AI in Infectious Diseases.
Lancet infographic summarising the key findings and recommendations from The Lancet Infectious Diseases Series on AI in Infectious Diseases.
Lancet infographic summarising the key findings and recommendations from The Lancet Infectious Diseases Series on AI in Infectious Diseases.
Lancet infographic summarising the key findings and recommendations from The Lancet Infectious Diseases Series on AI in Infectious Diseases.
The role of AI in infectious diseases is emerging—offering new opportunities for infection prevention, detection & control.
Explore developments & practical applications ⤵️
spkl.io/63323Ad83c @thelancetinfdis.bsky.social #WorldAMRAwarenessWeek
If successful for drug delivery, this could be really be helpful for minimizing adverse effects of treatment for infections… question is, can it get into a non-vascular space? (Ie abscesses or granulomas)…and can it be done cheaply so that we all can use it?
Thank you @tedpak.bsky.social for publishing the code for the LLM medical concept extractor you built for our recent paper! 2 things make this significant: a) This isn’t done enough in translational research and b) it’s designed for general use (ie Dockerized). Superb!
github.com/tpaklab/llacie
Our lab's paper describing the North American H5N1 epizootic is out now in Nature! So thrilled to have this out, and congratulations to @lambod50.bsky.social for all the fantastic work on this: www.nature.com/articles/s41...
Sorry for the delayed reply; been a while since I thought of this but yes my recollection is that since BAL samples are ~50mls, we centrifuge prior to Gram staining and aliquot 0.01ml for quantitative culture. Happy to inquire with folks on the bench if that would be of help!
Gave my 1st talk in NL: how to make multicenter data analysis with patient data actually work.
tl;dr: common data models (global interoperability) + LLMs (data harmonization, mapping, future-proofing) + federated learning (privacy)
Let me know what you think!
drive.google.com/file/d/1N3yk...
Last chance world! Application closes tomorrow!
