Tal Korem
@tkorem.bsky.social
Microbiome, metagenomics, ML, and reproductive health. All views are mine. So are all your base
Reposted by Tal Korem
Pollution dropping by that much likely means a whole bunch of people are alive who otherwise wouldn’t be. Incredible stuff.
Year 1 data on congestion pricing in Manhattan…
* Vehicle traffic: -11%
* Foot traffic: +3.4%
* Storefront vacancy: -0.9%
* Pollution: -22%
* Revenue for mass transit: $548M
So YES this has been a huge success.
* Vehicle traffic: -11%
* Foot traffic: +3.4%
* Storefront vacancy: -0.9%
* Pollution: -22%
* Revenue for mass transit: $548M
So YES this has been a huge success.
December 23, 2025 at 10:00 PM
Pollution dropping by that much likely means a whole bunch of people are alive who otherwise wouldn’t be. Incredible stuff.
Reposted by Tal Korem
We just published in @molsystbiol.org with the Mugler lab (UPitt) on bacterial population dynamics during tumor colonization (mouse model). Our study was guided by a Luria–Delbrück-style idea: infer mechanism from statistics (1/7) 🧪🦠
doi.org/10.1038/s443...
doi.org/10.1038/s443...
December 17, 2025 at 6:52 PM
We just published in @molsystbiol.org with the Mugler lab (UPitt) on bacterial population dynamics during tumor colonization (mouse model). Our study was guided by a Luria–Delbrück-style idea: infer mechanism from statistics (1/7) 🧪🦠
doi.org/10.1038/s443...
doi.org/10.1038/s443...
Reposted by Tal Korem
Out after peer-review: www.science.org/doi/full/10....
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
November 28, 2025 at 7:32 PM
Out after peer-review: www.science.org/doi/full/10....
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
Reposted by Tal Korem
Using leave-one-out cross-validation to calculate metrics such as AUC and R^2 creates bias! This can be fixed by removing one of each class in the meanwhile to maintain the training data distribution - great work by Tal and his team 🧪🧬🖥️
Out after peer-review: www.science.org/doi/full/10....
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
November 29, 2025 at 1:04 PM
Using leave-one-out cross-validation to calculate metrics such as AUC and R^2 creates bias! This can be fixed by removing one of each class in the meanwhile to maintain the training data distribution - great work by Tal and his team 🧪🧬🖥️
Reposted by Tal Korem
I found a flowchart which helps you navigate the IT landscape
October 1, 2025 at 6:22 PM
I found a flowchart which helps you navigate the IT landscape
Reposted by Tal Korem
Back when we hid beer in the cold room in a box labeled “yeast embryos.”
November 29, 2025 at 12:07 AM
Back when we hid beer in the cold room in a box labeled “yeast embryos.”
Out after peer-review: www.science.org/doi/full/10....
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
November 28, 2025 at 7:32 PM
Out after peer-review: www.science.org/doi/full/10....
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
Our bottom line stayed: never use leave-one-out cross-validation as it has inherent train-test leakage. Consider our Rebalanced version instead!
We now also account for regression and nested cross-validation, with more extensive benchmarking.
As long as three reviewers keep reading each proposal, doesn't really address anything either
November 13, 2025 at 11:23 PM
As long as three reviewers keep reading each proposal, doesn't really address anything either
Reposted by Tal Korem
"This could take down Democrats, too."
I know. And I frankly wouldn't give even an itty bitty damn if it implicated every Democratic man in Congress, every Democratic hopeful for 2028 and every Democrat who has even thought about running for office.
Down with the sex predators, wherever they are.
I know. And I frankly wouldn't give even an itty bitty damn if it implicated every Democratic man in Congress, every Democratic hopeful for 2028 and every Democrat who has even thought about running for office.
Down with the sex predators, wherever they are.
November 13, 2025 at 1:15 AM
"This could take down Democrats, too."
I know. And I frankly wouldn't give even an itty bitty damn if it implicated every Democratic man in Congress, every Democratic hopeful for 2028 and every Democrat who has even thought about running for office.
Down with the sex predators, wherever they are.
I know. And I frankly wouldn't give even an itty bitty damn if it implicated every Democratic man in Congress, every Democratic hopeful for 2028 and every Democrat who has even thought about running for office.
Down with the sex predators, wherever they are.
Reposted by Tal Korem
White paper: github.com/aqlaboratory...
GitHub Repo: github.com/aqlaboratory...
Huggingface: huggingface.co/OpenFold/Ope...
Testimonials: openfold.ghost.io/openfold3-co...
GitHub Repo: github.com/aqlaboratory...
Huggingface: huggingface.co/OpenFold/Ope...
Testimonials: openfold.ghost.io/openfold3-co...
github.com
October 28, 2025 at 6:30 PM
White paper: github.com/aqlaboratory...
GitHub Repo: github.com/aqlaboratory...
Huggingface: huggingface.co/OpenFold/Ope...
Testimonials: openfold.ghost.io/openfold3-co...
GitHub Repo: github.com/aqlaboratory...
Huggingface: huggingface.co/OpenFold/Ope...
Testimonials: openfold.ghost.io/openfold3-co...
Reposted by Tal Korem
OpenFold3-preview (OF3p) is out: a sneak peek of our AF3-based structure prediction model. Our aim for OF3 is full AF3-parity for every modality. We now believe we have a clear path towards this goal and are releasing OF3p to enable building in the OF3 ecosystem. More👇
October 28, 2025 at 6:30 PM
OpenFold3-preview (OF3p) is out: a sneak peek of our AF3-based structure prediction model. Our aim for OF3 is full AF3-parity for every modality. We now believe we have a clear path towards this goal and are releasing OF3p to enable building in the OF3 ecosystem. More👇
Reposted by Tal Korem
Just to be super clear, if you’re phoning in your peer review to ai you should quit your job so someone else who actually likes science can have it.
October 16, 2025 at 11:37 PM
Just to be super clear, if you’re phoning in your peer review to ai you should quit your job so someone else who actually likes science can have it.
No idea. Still working through this.
September 19, 2025 at 3:06 AM
No idea. Still working through this.
This is from the very paper you linked to - Figure S5. They claim that this has a p-value of 1. It's not here and there, this is what most results look like, and this is a large part of the basis for claiming that there are no robust associations with other tumors.
September 19, 2025 at 3:01 AM
This is from the very paper you linked to - Figure S5. They claim that this has a p-value of 1. It's not here and there, this is what most results look like, and this is a large part of the basis for claiming that there are no robust associations with other tumors.
So you look at this figure and your interpretation is "no signal"?
September 18, 2025 at 11:50 PM
So you look at this figure and your interpretation is "no signal"?
I never knew I needed this thread
The only visually overwhelming flag that is also a 10/10 state flag
Happy Friday to the Maryland State flag
August 16, 2025 at 2:48 AM
I never knew I needed this thread
We are hiring a postdoc - come work with us (www.koremlab.science) at the intersection of #microbiome, data science, and women's health!
Message or email me if interested. 🖥️ 🧬
Message or email me if interested. 🖥️ 🧬
Korem Lab - Microbiome Systems Biology @ Columbia | New York, USA
We apply systems biology approaches to decipher the metabolic interactions between the microbiome and its human host in diverse clinical settings, aiming towards personalized microbiome-based therapeu...
www.koremlab.science
August 7, 2025 at 1:41 PM
We are hiring a postdoc - come work with us (www.koremlab.science) at the intersection of #microbiome, data science, and women's health!
Message or email me if interested. 🖥️ 🧬
Message or email me if interested. 🖥️ 🧬
Reposted by Tal Korem
I am seeking a postdoc for my group at UCLA. We work at the intersection of population genetics x microbiome (garud.eeb.ucla.edu). If interested, please message me!
Garud Lab
garud.eeb.ucla.edu
July 22, 2025 at 5:51 PM
I am seeking a postdoc for my group at UCLA. We work at the intersection of population genetics x microbiome (garud.eeb.ucla.edu). If interested, please message me!
This will also likely reduce the number of study sections, firing SROs and making them less specialized.
July 17, 2025 at 9:51 PM
This will also likely reduce the number of study sections, firing SROs and making them less specialized.
Reposted by Tal Korem
Hope this is useful - consensus statement "Guidelines for preventing and reporting contamination in low-biomass microbiome studies" rdcu.be/er3Io
Guidelines for preventing and reporting contamination in low-biomass microbiome studies
Nature Microbiology - In this Consensus Statement, the authors outline strategies for processing, analysing and interpreting low-biomass microbiome samples, and provide recommendations to minimize...
rdcu.be
June 21, 2025 at 7:59 PM
Hope this is useful - consensus statement "Guidelines for preventing and reporting contamination in low-biomass microbiome studies" rdcu.be/er3Io
Reposted by Tal Korem
Reposted by Tal Korem
Not getting much attention, except in a recent NYTimes story, is a provision to increase the tax on university endowments and those of other nonprofits.
May 22, 2025 at 10:22 AM
Not getting much attention, except in a recent NYTimes story, is a provision to increase the tax on university endowments and those of other nonprofits.
Reposted by Tal Korem
New paper in Genome Biology!
genomebiology.biomedcentral.com/articles/10....
We introduce scale models, a generalization of normalizations that explciitly account for uncertainty in biological system scale (e.g., microbial load).
genomebiology.biomedcentral.com/articles/10....
We introduce scale models, a generalization of normalizations that explciitly account for uncertainty in biological system scale (e.g., microbial load).
Incorporating scale uncertainty in microbiome and gene expression analysis as an extension of normalization - Genome Biology
Statistical normalizations are used in differential analyses to address sample-to-sample variation in sequencing depth. Yet normalizations make strong, implicit assumptions about the scale of biologic...
genomebiology.biomedcentral.com
May 22, 2025 at 4:43 PM
New paper in Genome Biology!
genomebiology.biomedcentral.com/articles/10....
We introduce scale models, a generalization of normalizations that explciitly account for uncertainty in biological system scale (e.g., microbial load).
genomebiology.biomedcentral.com/articles/10....
We introduce scale models, a generalization of normalizations that explciitly account for uncertainty in biological system scale (e.g., microbial load).
Reposted by Tal Korem
as a resident of syracuse, ny, a rust belt town that used to be an economic epicenter for the nation: syracuse university is our largest local employer now and if it goes under, so does my town, which has the largest concentration of child poverty in the nation.
There is a false dichotomy drawn between "the ivory tower" and "the real world," and I'm here to report that in a post-industrial society, your real-world economy absolutely hinges on the university.
University towns are factory towns. Universities drive economic activity, not the other way around.
University towns are factory towns. Universities drive economic activity, not the other way around.
A must read
May 18, 2025 at 2:16 PM
as a resident of syracuse, ny, a rust belt town that used to be an economic epicenter for the nation: syracuse university is our largest local employer now and if it goes under, so does my town, which has the largest concentration of child poverty in the nation.
Reposted by Tal Korem
"I'm still in shock. I know that there have been political issues around Harvard in recent weeks, but antibiotic resistance isn't one of them." My conversation with Harvard microbiologist @baym.lol, one of many researchers there who just lost millions in fed. grants. www.wbur.org/news/2025/05...
Antibiotic research at Harvard lab threatened by federal funding cuts
Microbiologist Michael Baym studies antibiotic resistance at Harvard Medical School. He lost millions in federal funding this week.
www.wbur.org
May 16, 2025 at 10:59 PM
"I'm still in shock. I know that there have been political issues around Harvard in recent weeks, but antibiotic resistance isn't one of them." My conversation with Harvard microbiologist @baym.lol, one of many researchers there who just lost millions in fed. grants. www.wbur.org/news/2025/05...