Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Cell atlases and the developmental foundations of the phenotype.

PLoS computational biology·2026
Same author

Structural Changes in Gene Ontology Reveal Modular and Complex Representations of Biological Function.

Molecular biology and evolution·2025
Same author

Polygenic prediction and gene regulation networks.

Royal Society open science·2025
Same author

Disentangling protein metabolic costs in human cells and tissues.

PNAS nexus·2025
Same author

Design principles of multi-map variation in biological systems.

Physical biology·2024
Same author

The limitations of phenotype prediction in metabolism.

PLoS computational biology·2023

Related Experiment Video

Updated: Jul 1, 2026

Zebrafish In Situ Spinal Cord Preparation for Electrophysiological Recordings from Spinal Sensory and Motor Neurons
08:24

Zebrafish In Situ Spinal Cord Preparation for Electrophysiological Recordings from Spinal Sensory and Motor Neurons

Published on: April 18, 2017

A transformer-based language model reveals developmental constraint and network complexity during zebrafish

Juan F Poyatos1

  • 1National Museum of Natural Sciences (MNCN-CSIC), Madrid 28006, Spain.

PNAS Nexus
|June 30, 2026
PubMed
Summary

Zebrafish development shows a transient reorganization of gene regulatory networks during mid-embryogenesis, revealing developmental constraint is embedded in network topology. Language models like Zebraformer can uncover complex biological structures from single-cell data.

Keywords:
developmental hourglasssingle-cell transcriptomicstransformer modelszebrafish development

More Related Videos

Flat Mount Preparation for Observation and Analysis of Zebrafish Embryo Specimens Stained by Whole Mount In situ Hybridization
06:36

Flat Mount Preparation for Observation and Analysis of Zebrafish Embryo Specimens Stained by Whole Mount In situ Hybridization

Published on: July 17, 2014

Imaging and 3D Reconstruction of Cerebrovascular Structures in Embryonic Zebrafish
08:00

Imaging and 3D Reconstruction of Cerebrovascular Structures in Embryonic Zebrafish

Published on: April 22, 2014

Related Experiment Videos

Last Updated: Jul 1, 2026

Zebrafish In Situ Spinal Cord Preparation for Electrophysiological Recordings from Spinal Sensory and Motor Neurons
08:24

Zebrafish In Situ Spinal Cord Preparation for Electrophysiological Recordings from Spinal Sensory and Motor Neurons

Published on: April 18, 2017

Flat Mount Preparation for Observation and Analysis of Zebrafish Embryo Specimens Stained by Whole Mount In situ Hybridization
06:36

Flat Mount Preparation for Observation and Analysis of Zebrafish Embryo Specimens Stained by Whole Mount In situ Hybridization

Published on: July 17, 2014

Imaging and 3D Reconstruction of Cerebrovascular Structures in Embryonic Zebrafish
08:00

Imaging and 3D Reconstruction of Cerebrovascular Structures in Embryonic Zebrafish

Published on: April 22, 2014

Area of Science:

  • Developmental biology
  • Computational biology
  • Genomics

Background:

  • The developmental hourglass framework proposes mid-embryogenesis (phylotypic stage) is crucial for conserved development.
  • Understanding regulatory complexity and constraint in development is a key biological challenge.

Purpose of the Study:

  • To test the developmental hourglass hypothesis using a novel language model.
  • To investigate the role of regulatory architecture in developmental constraint during the phylotypic stage.

Main Methods:

  • Trained Zebraformer, a transformer-based language model, on zebrafish single-cell transcriptomic data.
  • Analyzed gene regulatory networks and their topological properties during embryogenesis.
  • Utilized graph-theoretic metrics and gene ontology enrichment analyses.

Main Results:

  • Zebraformer learned representations capturing temporal progression and anatomical identity.
  • Identified a transient reorganization of gene regulatory networks during the phylotypic stage, with coordinated modules and reduced connectivity.
  • Demonstrated that developmental constraint is linked to network topology, not just perturbation magnitude.

Conclusions:

  • Refined the hourglass framework by localizing developmental constraint to gene regulatory network architecture.
  • Showcased the utility of language models for extracting interpretable biological structures from single-cell data.
  • Highlighted the importance of network topology in understanding developmental constraint.