Sign in

Shiladitya Banerjee

@shilabanerjee.bsky.social
726 followers 265 following 33 posts

Living Matter Theory @ Georgia Tech shiladitya-banerjee.com

PostsRepliesMedia
Reposted by Shiladitya Banerjee
The Conversation U.S. @us.theconversation.com · 12/09/2026
🦠‼️ Bacteria can store memories, learn from past experiences and use that information to prepare for the future, according to a new study. And they may be doing it using some of the same basic logic as #AI. theconversation.com/learning-wit...
theconversation.com
Learning without a brain – how bacteria store memories and remember the past like artificial neural networks
Bacteria use strategies similar to artificial neural networks to use the past to adapt to the future.
32417
Reposted by Shiladitya Banerjee
Yanlan Mao @yanlanmao.bsky.social · 30/07/2026
This epic paper, which has been a heroic 10 year journey for the lead author, Ricardo Barrientos, is finally out in Cell today! Many thanks to all co-authors, especially Billie Meadowcroft and Andela Saric for their beautiful MD simulations www.cell.com/cell/fulltex...
cell.com
Basement membrane turnover controls cell shape
Basement membranes are scaffolds of proteins, like collagen, that give organs shape. During development, basement membranes must adapt to the growing organ. We infer how fast the basement membrane ada...
112654
Shiladitya Banerjee @shilabanerjee.bsky.social · 03/08/2026
experimental sciences will prevail and become more popular majors
010
Shiladitya Banerjee @shilabanerjee.bsky.social · 02/08/2026
Unhealthy North Indian food
000
Shiladitya Banerjee @shilabanerjee.bsky.social · 21/07/2026
Congratulations!
000
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/05/2026
9/ Bottom line: history-dependent computation does not require neurons. A reaction network with the right architecture is enough, and bacteria have been running this algorithm for billions of years. Work with Huijing Wang, @sifangwei.bsky.social, led by Josiah Kratz.
0130
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/05/2026
8/ The reaction network governing growth is, mathematically, a continuous-time RNN. Ribosomal sub-sectors are the hidden state. Nutrient-dependent allocation is the gating mechanism, the same trick LSTMs use to balance memory and flexibility. Evolution and ML converged on the same architecture.
1110
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/05/2026
7/ There's a cost to adaptation. Cells that adapt fast grow slower on average. We see this in the data: long-term growth rate and adaptation timescale are positively correlated. Faster learning costs steady-state fitness. Cells appear to actively navigate this tradeoff.
1130
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/05/2026
6/ Where does scale-free memory come from in a cell? We propose that it emerges from ribosomes themselves. Ribosomes are regulated on many timescales (ppGpp, hibernation factors, degradation). A power-law distribution of relaxation rates produces exactly the fractional dynamics we observe.
1121
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/05/2026
5/ Next came the real test. We trained cells at one pulsing frequency, then abruptly switched to a faster one. With no refitting, the model predicted how quickly cells would re-tune. Two pulses and they've adjusted, in agreement with experiments.
1121
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/05/2026
4/ Standard proteome allocation models (Markovian, one timescale) cannot reproduce this. They predict cells should fully reset between pulses. Real cells don't. So we rewrote the dynamics using fractional calculus, which has memory built in. The new model quantitatively explains the data.
1200
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/05/2026
3/ Answer: They use the pattern. The time a cell takes to ramp up growth after a nutrient pulse depends on the history of pulses it has seen. And the relationship between fluctuation frequency and adaptation speed follows a power law. No single timescale. Memory across minutes to hours.
1100
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/05/2026
2/ @sifangwei.bsky.social tracked individual E. coli in a microfluidic device while switching nutrients on and off at different frequencies. The question: do cells just simply respond to the present environmental condition, or do they use the pattern of past nutrients to prepare for what's next?
1130
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/05/2026
1/ Can a single bacterium learn? No brain, no neurons, just one cell deciding what to do next. Our new paper says yes, and the cell's internal machinery turns out to compute exactly like a recurrent neural network. 🔗 journals.aps.org/prxlife/abstract/10.1103/5zbg-8vll
2426
Reposted by Shiladitya Banerjee
Georgia Tech Research @gtresearch.bsky.social · 28/04/2026
In about one out of every 1,000 pregnancies, the neural tube, a key nervous system structure, fails to close. @gtsciences.bsky.social is now helping explain why this happens, having uncovered the physics that drive neural tube closure in a pregnancy’s earliest stages. b.gatech.edu/421VdDG
The neural tube
022
Shiladitya Banerjee @shilabanerjee.bsky.social · 25/03/2026
Check our new paper with @glgalea.bsky.social on the mechanics of hindbrain neuropore morphogenesis www.cell.com/current-biol...
cell.com
Mechanosensitive feedback organizes cell shape and motion during hindbrain neuropore morphogenesis
Pérez-Verdugo et al. show that hindbrain neuropore closure is driven by surface ectoderm actomyosin purse-string tension and active migration. These forces create reproducible patterns of cell elongat...
031
Reposted by Shiladitya Banerjee
Sadjad Arzash @sadjadarzash.bsky.social · 09/03/2026
Excited to present this work w/ @shilabanerjee.bsky.social next week at APS Global Physics Summit 2026 on mechanical memory and learning in biological tissues. bioRxiv: www.biorxiv.org/content/10.6...
041
Shiladitya Banerjee @shilabanerjee.bsky.social · 20/12/2025
Can biological tissues "learn"? Our new preprint shows that epithelial tissues exhibit emergent behaviors akin to unsupervised learning. Local tension remodeling allows cell networks to store long-range memory and program global elasticity properties. Read more here: www.biorxiv.org/content/10.6...
0318
Shiladitya Banerjee @shilabanerjee.bsky.social · 25/11/2025
www.biorxiv.org/content/10.1...
biorxiv.org
Reinforcement learning for adaptive control of phenotypically heterogeneous bacterial populations
Bacterial populations display extraordinary resilience to antibiotic stress, driven by diverse physiological states that allow some cells to persist and later repopulate. This phenotypic heterogeneity...
020
Shiladitya Banerjee @shilabanerjee.bsky.social · 09/10/2025
royalsocietypublishing.org/doi/10.1098/...
royalsocietypublishing.org
Physical traits of supercompetitors in cell competition | Journal of The Royal Society Interface
Cell competition is a fitness control mechanism in tissues, where less fit cells are eliminated to maintain tissue homeostasis. Two primary mechanisms of cell competition have been identified: contact...
020
Shiladitya Banerjee @shilabanerjee.bsky.social · 15/08/2025
Cell Geometry Limits Metabolic Efficiency www.biorxiv.org/content/10.1...
biorxiv.org
Cell Geometry Limits Bacterial Metabolic Efficiency
Bacterial metabolic strategies are fundamentally linked to their physical form, yet a quantitative understanding of how cell size and shape constrain the efficiency of biomass production remains poorl...
030
Shiladitya Banerjee @shilabanerjee.bsky.social · 22/05/2025
Check out our new opinion/review on Organelle Size Control www.sciencedirect.com/science/arti...
sciencedirect.com
Design principles and feedback mechanisms in organelle size control
Intracellular organelles are essential for cellular architecture and function, and their size regulation is critical for maintaining cellular homeosta…
051
Shiladitya Banerjee @shilabanerjee.bsky.social · 09/05/2025
That's excellent!
010
Shiladitya Banerjee @shilabanerjee.bsky.social · 30/04/2025
Excited to share our new preprint on bacterial learning, with @sifangwei.bsky.social. Here we show that single bacterial cells can continually learn in dynamic environments, due to an underlying reaction network architecture that resemble a recurrent neural network www.biorxiv.org/content/10.1...
biorxiv.org
Power-law memory governs bacterial adaptation and learning in fluctuating environments
How do single-celled organisms adapt and learn to survive in dynamic environments without a nervous system? Here, we provide experimental evidence and a theoretical model demonstrating learning-like b...
081
Shiladitya Banerjee @shilabanerjee.bsky.social · 08/04/2025
good journal. Congrats!
110
Shiladitya Banerjee @shilabanerjee.bsky.social · 04/04/2025
010
Shiladitya Banerjee @shilabanerjee.bsky.social · 02/04/2025
Congrats Jake
010
Reposted by Shiladitya Banerjee
PoLS at Georgia Tech @gtpols.bsky.social · 27/03/2025
PoLS faculty Dan Goldman has been named a lifetime fellow of the American Association for the Advancement of #Science (@aaas.org)! Congratulations to Dan, and also to the six other @gtresearchnews.bsky.social honorees named this year!
143
Shiladitya Banerjee @shilabanerjee.bsky.social · 21/02/2025
Read about our latest work on centrosome size control, Version of Record now available @elife.bsky.social
043
Reposted by Shiladitya Banerjee
Ulrich Schwarz @ulrichschwarz.bsky.social · 17/02/2025
Big shout out to Jian Liu, @shilabanerjee.bsky.social and @simonereber.bsky.social for organizing a great subgroup meeting on Physical Cell Biology at #BPS2025 at Los Angeles. And afterwards a very enjoyable lunch for the speakers😁
0151
Shiladitya Banerjee @shilabanerjee.bsky.social · 12/02/2025
Beautiful work. Congrats!
100
Shiladitya Banerjee @shilabanerjee.bsky.social · 11/02/2025
www.aamc.org/news/press-r...
aamc.org
AAMC Lawsuit Results in Nationwide Temporary Restraining Order on Proposed Drastic Cuts to NIH Funding
Judge grants restraining order after lawsuit led by the AAMC, challenging the legality of the administration’s plan to make cuts to federally funded research.
010
Reposted by Shiladitya Banerjee
PoLS at Georgia Tech @gtpols.bsky.social · 07/02/2025
How can subterranean organisms be observed as they move underground? A recent paper from Dan Goldman's CRAB Lab presents Laser Speckle Imaging (LSI) as a non-invasive way to view organisms as they move through granular media. #Physics #Biology doi.org/10.1242/jeb....
doi.org
Detecting subtle subterranean movement via laser speckle imaging
Summary: Laser speckle imaging non-invasively quantifies subterranean organism activity by detecting nanometer-scale movements.
111
Shiladitya Banerjee @shilabanerjee.bsky.social · 23/12/2024
Hi Folks, the Physical Cell Biology subgroup of the BPS is currently searching for a secretary treasurer. We are seeking two nominations to run for an election. DM me if interested. Pleaee Repost.
000
Shiladitya Banerjee @shilabanerjee.bsky.social · 12/12/2024
A betrayal to those who opted for the new framework. Why not simply accept/reject then?
000
Shiladitya Banerjee @shilabanerjee.bsky.social · 08/12/2024
Please cite doi.org/10.7554/eLif... as you are using a movie from this paper without crediting
doi.org
Cell-scale biophysical determinants of cell competition in epithelia
Mechanical competition is controlled by biophysical parameters that regulate cellular homeostatic density, while biochemical competition is regulated by parameters that influence the organisation of c...
070
Shiladitya Banerjee @shilabanerjee.bsky.social · 22/11/2024
Work led by postdoc Fernanda Perez-Verdugo, with a very productive collaboration with Gabriel Galea's lab @UCL and @emaniou.bsky.social. Many thanks to Gabe for introducing me to this intriguing model system. Read our earlier work on hindbrain neuropore closure here: www.pnas.org/doi/abs/10.1...
pnas.org
Hindbrain neuropore tissue geometry determines asymmetric cell-mediated closure dynamics in mouse embryos | PNAS
Gap closure is a common morphogenetic process. In mammals, failure to close the embryonic hindbrain neuropore (HNP) gap causes fatal anencephaly. W...
010
Shiladitya Banerjee @shilabanerjee.bsky.social · 22/11/2024
How do physical forces influence spatiotemporal cell patterning during tissue morphogenesis? Our latest preprint integrates theoretical modeling with live imaging of mouse embryos to reveal how mechanosensitive feedback drives self-organized cell patterning during hindbrain neuropore morphogenesis.
23510
Shiladitya Banerjee @shilabanerjee.bsky.social · 20/11/2024
How about 'Blueps'
010
Shiladitya Banerjee @shilabanerjee.bsky.social · 17/11/2024
🎉 Excited to share the news that my group will soon be moving to Georgia Tech to begin a new chapter in the Physics of Living Systems. Grateful for an incredible 5 years at CMU. We’re hiring postdocs - join us!
3180
Reposted by Shiladitya Banerjee
Kirsty Wan @micromotility.bsky.social · 13/11/2024
Couldn't find one so made a quick physics of life list, enjoy! go.bsky.app/EA2VsDN
7813077