Alpha1 Record ID: 51C95397
3 views
Share this Alpha1 record with your network
Top 1% Selection Manuscript
Loading
Fetching the latest research
Nener, J.; Selvamani, P.; Badarinarayan, S. S.; Chandramohan, N.; Grzybowski, A. T.
Single-cell foundation models are reshaping biology by learning transferable representations of cellular state from millions of profiles. These models support annotation, denoising, cross-modal mapping and, increasingly, prediction of responses to genetic or pharmacological perturbations. Despite this progress, most approaches treat cells as independent observations and ignore the multicellular context that governs tissue behavior. Models trained on aggregated datasets often fail to generalize to new donors, laboratories or interventions, in part because their latent spaces lack structure for composition and extrapolation. As a result, strong reconstruction performance does not guarantee accurate forecasting of system-level responses. The general problem addressed here is how to construct a scalable model that predicts multicellular, patient-level responses to interventions while preserving single-cell resolution and enabling generalization beyond observed conditions. Here we show that TRAILBLAZER, a multicellular transformer encoder coupled to an explicitly shaped hyperspherical latent space and a count-aware generative decoder, enables accurate zero-shot prediction of perturbation responses and ranking of candidate immunomodulators at patient resolution. In contrast to prior single-cell or pseudo-bulk approaches, TRAILBLAZER models tissues as coordinated systems using latent tokens that summarize and redistribute global context while maintaining near-linear scaling with group size. By organizing latent geometry around shared healthy references and calibrated mechanistic directions, the model renders vector arithmetic biologically meaningful and supports extrapolation to unseen agents. Together, these results establish a practical framework for mechanism-aware simulation of multicellular responses and suggest a path toward predictive foundation models for therapeutic discovery.
TRAILBLAZER is a transformer that models entire multicellular neighborhoods at once, predicting how patient cell populations will react to drugs in zero-shot—think a hyperspherical crystal ball that ranks immunomodulators like a quirky matchmaker for your immune system.
Noted for its innovative multicellular generative approach and patient-level predictions, sparking interest among systems biologists working on perturbation modeling
View discussionFormal Expert Review in progress...
Read the Kaimen Rigor review above, then add your own assessment.
CD4⁺ T cells confer transplantable rejuvenation via Rivers of telomeres
Lanna, A.; Valvo, S.; Dustin, M.; Rinaldi, F.
Using a GPT-5-driven autonomous lab to optimize the cost and titer of cell-free protein synthesis
Smith, A. A.; Wong, E. L.; Donovan, R. C.; Chapman, B. A.; Harry, R.; Tirandazi, P.; Kanigowska, P.; Gendreau, E. A.; Dahl, R. H.; Jastrzebski, M.; Cortez, J. E.; Bremner, C. J.; Hemuda, J. C. M.; Dooner, J.; Graves, I.; Karandikar, R.; Lionetti, C.; Christopher, K.; Consiglio, A. L.; Tran, A.; McCusker, W.; Nguyen, D. X.; Nunes da Silva, I. B.; Bautista-Ayala, A. R.; McNerney, M. P.; Atkins, S.; McDuffie, M.; Serber, W.; Barber, B. P.; Thanongsinh, T.; Nesson, A.; Lama, B.; Nichols, B.; LaFrance, C.; Nyima, T.; Byrn, A.; Thornhill, R.; Cai, B.; Ayala-Valdez, L.; Wong, A.; Che, A. J.; Thavaraj
A Single-Cell and Spatial 3D Multi-omic Atlas of Developing Human Basal Ganglia and Inhibitory Neurons
Heffel, M. G.; Xu, H.; Pastor-Alonso, O.; Li, X.; Baig, M. S.; Irfan Ghoor, R.; Li, R.; Kern, C.; Kum, J.; Zhang, Y.; Paino, J.; Tsai, M. J.; Tai, C.-Y.; Tucker, G.; Zhao, Z.; Hou, A.; von Behren, Z.; Bhade, M.; Li, S.; Sandoval, K.; Scholes, J.; Codrea, F.; Calimlim, J.; Liao, E. K.; Leung, G.; Kim, J.; Eskin, E.; Flint, J.; Cotter, J. A.; Pasaniuc, B.; Bintu, B.; Zhu, Q.; Mukamel, E. A.; Ernst, J.; Paredes, M. F.; Luo, C.
Prediction of transformative breakthroughs in biomedical research
Davis, M. T.; Busse, B. L.; Arabi, S.; Meyer, P.; Hoppe, T. A.; Meseroll, R. A.; Hutchins, B. I.; Willis, K. A.; Santangelo, G. M.