CELL @CellBioSF

CELL: Consortium for the Equations of Life and Living Systems. Fusing MathBio, BioPhysics, CompBio and DescriptiveBio around an aggressive mathematical core. cellbiosf.substack.com San Francisco Joined August 2025

Tweets

426
Followers

535
Following

225
Likes

417

Joel David Hamkins @JDHamkins

2 days ago

Next week's reading for the core PhilSciMath seminar, two sessions.

15 81 926 34K 500

Download Image

Thinking Machines @thinkymachines

2 days ago

Efficient training of neural networks is difficult. Our second Connectionism post introduces Modular Manifolds, a theoretical step toward more stable and performant training by co-designing neural net optimizers with manifold constraints on weight matrices.…

108 436 3K 1.2M 2K

Download Image

Sam Altman @sama

3 days ago

Today we are launching my favorite feature of ChatGPT so far, called Pulse. It is initially available to Pro subscribers. Pulse works for you overnight, and keeps thinking about your interests, your connected data, your recent chats, and more. Every morning, you get a…

3K 3K 42K 6.7M 11K

Math Cafe @Riazi_Cafe_en

3 days ago

Cambridge "Statistics" lecture notes PDF: statslab.cam.ac.uk/~rrw1/stats/Sa… Example sheets & more: statslab.cam.ac.uk/~rrw1/stats/in…

2 116 765 27K 822

Download Image

Anshul Kundaje (anshulkundaje@bluesky) @anshulkundaje

5 days ago

2 cool papers on sequence-to-gene expression models in yeast biorxiv.org/content/10.110… (pretrains a fungal DNALM -> fine tunes on yeast expression & ChIP-exo profiles) biorxiv.org/content/10.110… (directly trains on expression profiles) Both use modified Borzoi architectures 1/

2 10 100 8K 52

Santiago Mille @santimillef

4 days ago

@_jnwang @talaldotpdb @driscoll_cl @haoyu_dai_dhy @Xiaowei0402 @SynBioGaoLab @BrianHie @sokrypton @richardwshuai @MartinPacesa We’re excited for a new era of antibody (and protein) design, and can’t wait to see how others use Germinal. 🚀 Read more in our preprint below, and reach out if you have questions or ideas—we’d love to hear from you! Code: github.com/SantiagoMille/… Preprint:…

2 10 45 4K 27

CELL @CellBioSF

5 days ago

Why a mathematical configuration space for biology? Please read.

CELL @CellBioSF

3 weeks ago

Why a mathematical configuration space for biology? Please read.

1 5 14 1K 5

Download Image

0 0 0 271 0

NSF-Simons NITMB @NITMB_Chicago

2 weeks ago

Are you interested in exploring your work from a new perspective with mathematical biology? Join us for our upcoming workshop, Expanding the Palette of Mathematics in Biology! Apply today at nitmb.org/expanding-pale…

0 5 11 1K 4

Download Image

DSABNS @dsabns

7 days ago

📢📢📢 Young Researcher Fellowship – DSABNS 2026 (Sponsored by SMB) The Society for Mathematical Biology (SMB) will support young researchers with 8 fellowships (covering the €250 early-bird registration fee) to attend the 17th DSABNS 2026 in Granada, Spain.

1 2 3 204 0

Probability and Statistics @probnstat

5 days ago

Empirical processes study how well observed data represents the true underlying distribution. In statistics, they form the basis for goodness-of-fit tests (e.g., Kolmogorov-Smirnov). In machine learning, they are the core of statistical learning theory, explaining model…

4 30 228 11K 139

Download Image

Yaashaa Golovanov @Golovanov_ammoc

5 days ago

Once Prof. Steven Strogatz had asked for some nice references on “the Principle of Least Action.” I am adding another one to the list.

Yaashaa Golovanov @Golovanov_ammoc

7 months ago

Once Prof. Steven Strogatz had asked for some nice references on “the Principle of Least Action.” I am adding another one to the list. https://t.co/nTmBH3WazP

3 38 190 43K 130

Download Image

3 83 675 34K 582

Download Image

CELL @CellBioSF

5 days ago

Here are 3 compelling reasons why Member Labs are joining CELL: 1) Work in one shared mathematical space: Put your cell types, assays, and perturbations into a common configuration space so models, datasets, and results plug together across projects and scales. Instead of…

0 1 1 270 3

Freedom Preetham @freedompreetham

6 days ago

Read this paper by @yun_s_song and team. Fasinating approach and insights. GPN-Star reframes genomic language modeling around 'evolution' rather than 'single-sequence context'. The core move is to treat a whole-genome alignment (WGA) plus a species tree as the input, compress…

Yun S. Song @yun_s_song

6 days ago

17 156 505 72K 280

Download Image

0 2 5 957 4

Yun S. Song @yun_s_song

6 days ago

To show that GPN-Star is a robust and generalizable framework that can advance biology beyond human genetics, we apply it to train gLMs for five well-studied model organisms and demonstrate their effectiveness in assessing variant effects in these species. (8/n)

1 1 11 2K 1

Download Image

Yun S. Song @yun_s_song

6 days ago

In addition, GPN-Star exhibits meaningful nucleotide dependencies that align with known functional dependencies, indicating its potential to help understand genomic syntax. This represents a notable advance over traditional conservation scores. (7/n)

1 2 14 2K 1

Download Image

Yun S. Song @yun_s_song

6 days ago

By training GPN-Star on vertebrate, mammal, and primate alignments, we reveal task-dependent advantages of modeling deeper versus more recent evolution. These findings offer new biological insights and practical guidance for developing future gLMs and evolutionary models. (6/n)

1 1 11 2K 0

Download Image

Yun S. Song @yun_s_song

6 days ago

GPN-Star achieves unprecedented SNP heritability enrichments across over 100 human complex traits. Moreover, we devise a simple approach to incorporate tissue-specificity into the model prediction and show that it further improves heritability enrichment. (5/n)

1 6 38 13K 10

Download Image

Yun S. Song @yun_s_song

6 days ago

We compare GPN-Star with several models, including the recent AlphaGenome and Evo2 models with up to 1Mb context size and 40B parameters, and observe that GPN-Star consistently ranks at the top across a wide range of human variant effect prediction tasks. (4/n)

1 2 22 3K 2

Download Image

Yun S. Song @yun_s_song

6 days ago

GPN-Star features a novel phylogeny-aware architecture that enables the model to explicitly capture evolutionary relationships encoded in whole-genome alignments and overcomes the key limitations of our earlier model GPN-MSA (doi.org/10.1038/s41587…). (2/n)