bioRxiv Science⌕ Search

Biology subjects

Dokmanic, I.

Publications and source records attributed to Dokmanic, I..

2 recordsLinked to original sources

Rapid canalization of chromosome conformation-transcription fingerprints during embryogenesis revealed by fully-automated cell identity decoding with CeSCALE

Genome organization into higher-order active and inactive compartments exhibits cell-type specific patterns, which are widely implicated in the regulation of transcriptional activity. During embryogenesis, epigenetic regulation controls cell type specification along cellular lineages with similar transcriptional identities through the coordinated action of chromatin states. However, prevalent single-molecule variability in higher-order chromosome conformation and a lack of precise cell lineage information have previously limited our understanding of the relationship between conformation and transcriptional activity in vivo. Specifically, how the conformation-transcription relationship is inherited along cellular lineages through cell divisions is poorly understood. Here, we developed a novel algorithm for cell lineage identification (C. elegans Sinkhorn-based Cell ALignmEnt, CeSCALE) combined with single cell genomics to reveal that local conformation-transcription fingerprints are associated with, and inherited along the stereotyped cellular lineages of C. elegans embryos. Inspired by Optimal Transport theory, CeSCALE provides a fully automated framework for quantifying lineage-resolved individual cell phenotypes in situ, across a wide developmental window. Combining CeSCALE with single-molecule chromosome tracing uncovered higher-order interchromosomal block associations, which surprisingly coalesce transcriptionally diverse domains and are independent of lineage identity. Instead, by integrating lineage-resolved chromosome conformations with single-cell transcriptomics, we find that local conformation-transcription spatial relationships ( fingerprints), containing both hubs and islands of transcriptional activity, are robustly inherited along lineages. Finally, we find that the canalization of these fingerprints represent the rewiring of chromatin states at key developmental stages. Our results suggest that local chromatin environments, but not large-scale compartments, coordinate the dramatically changing transcriptome during embryogenesis.

developmental biology↗

Icecream: High-Fidelity Equivariant Cryo-Electron Tomography

Cryo-electron tomography (cryo-ET) visualizes 3D cellular architecture in near-native states. Recent deep-learning methods (CryoCARE, IsoNet, DeepDeWedge, CryoLithe) improve denoising and artifact correction, but performance remains limited by very low signal-to-noise ratio, a restricted angular range ( missing wedge), and the lack of ground truth. Here, we present Icecream, which follows the broad template of earlier self-supervised approaches, but treats symmetry in a way consistent with the recent equivariant imaging framework (Chen et al., 2021). Coupled with several engineering refinements, including mixed-precision arithmetic, Icecream achieves substantially better denoising and more reliable missing-wedge recovery, while reducing training and inference time relative to comparable baselines. Across diverse experimental datasets, we observe consistent gains in reconstruction quality, both visually and as quantified by Fourier shell correlation (FSC). Our framework extends to any tomography problem that provides two statistically independent reconstructions of the same volume; in cryo-ET these are obtained by dose splitting or angular partitioning of the tilt series.

bioinformatics↗