ng 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 mech

atient-level respons

ntext that governs tissue behavior. Models trained on aggregated datasets often fail to generalize to new don

xiv thank the following for their generous financial support:The Chan Zuckerberg Initiative, Cold Spring Harbor Laboratory, the Sergey Brin Family Foundation, California Institute of Technology, Centre National de la Recherche Scientifique, Fred Hutchinson Cancer Center, Imper

ed to be regulated by post-translational modifications. Here, we show that phosphorylati

nized to be regulated by post-translational modifications. Here, we show that phosphorylation of a conserved threonine (T495), previously shown to be e

cal pho

Awardee in Aging and Biology and Geroscience Copyright The cop

ared imaging of fatty acid esterification in the ER of living cells

IR) microspectroscopy to investigate palmitic acid (PA) metabolism in hepatocytes with sub-micron spatial resolution. Upon PA feeding, we discover a time-dependent ester carbonyl stretch localized to the ER and lipid droplet-ER contact sites of lipid droplets with abnormal

Funder Information DeclaredAmerican Federation f

rting a model whereby PA acyl chain packing in the ER reduces enzyme diffusion slowing PA metabolism. Our results provide a deeper understanding of how phase changes induced by h

ges visualizing reactions in live cells. Here, we use optical photothermal infrared (OPTIR) microspectroscopy to investigate palmitic acid (PA) metabolism in hepatocytes with sub-micron spatial resolution. Upon PA feeding, we discover a time-dependent ester carbonyl stretch localized to the ER and lipid droplet-ER contact sites of lipid droplets with abnormal morphology. This s

llatio

te dapibus diam. Sed nisi. Nulla quis sem at nibh elementumimperdiet. Duis sagittis ipsum. Praesent mauris. Fusce nec tellus sed augue semper porta. Maurismassa. Vestibulum lacinia arcu eget nulla. Class aptent taciti sociosqu ad litora torquent perconubia nostra, per inceptos himenaeos.

eclared no competing interest.FootnotesText has been modified for clarity. There have been no changes to the data presented. Copyright The copyright holder fo

rims are still formed in a knockin mouse in which the tetraspanin body of PRPH2 was replaced wit

dy, we found that the knockout of ROM1 causes a compensatory increase in the disc content of PRPH2. Despite this increase, discs of ROM1 knockout mice displayed a delay in disc enclosure as

further increasing the level of PRPH2 rescued these morphological defects. We next showed that disc rims are still formed in a knockin mouse in which the tetraspanin body of PRPH2 was replaced with that of ROM1. Together, these results demonstrate that, despite its contribution t

anin proteins, PRPH2 (a.k.a. peripherin-2 or rds) and ROM1. While mutations in PRPH2 affect the formation of disc rims, the role of ROM1 remains poorly understood. In this study, we found th

eralization 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 pertu

llular 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 no

nformation DeclaredNIH Common Fund, https://ror.org/001d55x84, R01NS059690, R01GM140440, R01GM144378Pew Charitable Trusts, A129837

phosphonull Hsp70 variants disrupt G1/S progression under normal and DNA-damaging conditions. Biochemically, the phosphomimetic T495E mutation locks Hsp70 in an open-like conformation without blocking substrate engagement. Together, our results reveal a conserved mechanism by which dynami

eed and scalability. Here we introduce Variational Inference with Node Embeddings (Vine), a computational method that combines an embedding of taxa in a high-dimensional space and a distance-based “decoder” with several algorithmic innovations to dramatically improve phylogene