What happened
Single-cell RNA sequencing measures cellular states at high resolution, but sparse high-dimensional count data remain difficult to model interpretably. We introduce the Single-Cell Euclidean Network Embedding (SCENE), a probabilistic latent-distance model that jointly embeds cells and genes from Unique Molecular Identifier (UMI) counts. SCENE treats the count matrix as a weighted bipartite cell-gene graph, where Euclidean distances represent transcriptional affinity, and combines this geometry with a zero-inflated count likelihood that separates gene detection from expression magnitude. Across real and simulated scRNA-seq datasets, SCENE recovers biologically structured cell and gene embeddings with state-of-the-art performance. Surprisingly, major biological structure is preserved in native two- and three-dimensional latent spaces, enabling directly interpretable visualization. Perturbation analyses show that SCENE organizes glucocorticoid-response genes and T-cell receptor regulatory programs coherently in gene space, capturing biology beyond cell-type separation. SCENE provides a transparent representation learning framework in which low-dimensional Euclidean geometry supports accurate modeling and biological interpretation. Update (2026-09-20): Learning interpretable kinetic models for biomolecular interaction networks . Many cellular machines operate through weak, transient, and multivalent interactions whose functional states are governed by recurring interaction patterns rather than persistent molecular geometries. Here, we introduce interaction-based Markov state models (iMSMs), which construct interpretable kinetic models using unsupervised clustering of time-averaged, identity-resolved interaction distributions around a focal entity. Applied to nucleocytoplasmic transport at two molecular resolutions, iMSMs resolve graded interaction states spanning strong, partial, and weak engagement. During pore transport, partially engaged states provide faster routes to disengagement than strongly bound states, while the networks reveal transport pathways, interaction hubs, bottlenecks, and kinetic commitment. At a finer resolution, FG motifs exchange contacts while remaining associated, recovering established slide-and-exchange dynamics. iMSMs reproduce free energies, permeabilities, and multi-step kinetics, with nearly fourfold faster permeability convergence from truncated trajectories than direct Update (2026-09-21): Gravlax: an annotation-independent molecular evidence archive for single-cell RNA-seq . A cell-by-gene count matrix is the artifact of a single-cell RNA-seq experiment that is most often stored, shared, and reanalyzed. It is the output of a computation whose inputs are the sequenced molecules and a gene annotation, and while the molecules never change, the annotation is revised continually. Once the matrix has been produced, the evidence behind it can no longer be reinterpreted. Recovering that evidence means returning to raw reads or alignments that are large, costly to process, and frequently unavailable. We ask whether a concise representation of the molecules themselves can be extracted once and reused indefinitely, to quantify under any future annotation, to query and discover features that no annotation yet describes, and to pool evidence across cells and samples. Our starting observation is that the procedures that turn alignments into counts, gene assignment and UMI collapse, never read most of what an alignment file contains. They consume relations among molecules such as shared genomic geometry, shared placements, barcode identity, and the equal