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

DNA Base Pairing02:27

DNA Base Pairing

33.0K
Erwin Chargaff’s rules on DNA equivalence paved the way for the discovery of base pairing in DNA. Chargaff’s rules state that in a double-stranded DNA molecule,
33.0K
DNA Base Pairing02:27

DNA Base Pairing

31.9K
31.9K
VSEPR Theory and the Effect of Lone Pairs04:01

VSEPR Theory and the Effect of Lone Pairs

52.7K
Effect of Lone Pairs of Electrons on Molecule Geometry
52.7K
Frequency-dependent Selection01:21

Frequency-dependent Selection

23.1K
When the fitness of a trait is influenced by how common it is (i.e., its frequency) relative to different traits within a population, this is referred to as frequency-dependent selection. Frequency-dependent selection may occur between species or within a single species. This type of selection can either be positive—with more common phenotypes having higher fitness—or negative, with rarer phenotypes conferring increased fitness.
23.1K
Sign Test for Matched Pairs01:17

Sign Test for Matched Pairs

386
The sign test for matched pairs offers a robust method for comparing two paired samples, often for the effects of an intervention in one of them. This method is very useful in situations where the underlying distribution of the data is unknown. The test compares two related samples—often pre- and post-treatment measurements on the same subjects—to determine if there are significant differences in their median values.
To conduct the sign test, we first calculate the differences in...
386
Maturation of Endosomes01:28

Maturation of Endosomes

5.7K
The early endosome containing internalized molecules matures through transformations in its location, morphology, intraluminal pH, and membrane protein composition. Together, these changes result in a more acidic late endosome that contains multiple intraluminal vesicles; therefore, the late endosome is also called a multivesicular body (MVB).
Changes in location
The maturing endosome moves along microtubules from the periphery of the cell towards the perinuclear region. This movement of the...
5.7K

You might also read

Related Articles

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

Sort by
Same author

Structural basis of Mlc-mediated transcriptional regulation of carbohydrate metabolism.

Nature communications·2026
Same author

From unsupervised selection of AIMs to likelihood-based BGA inference: Challenges revealed by OpenADMIXTURE and GENOGEOGRAPHER.

Forensic science international. Genetics·2026
Same author

Beyond performance: how design choices shape chemical language models.

Journal of cheminformatics·2025
Same author

Tracking coordinated cellular dynamics in time-lapse microscopy with ARCOS.px.

Journal of cell science·2025
Same author

Structural basis of drug recognition by human MATE1 transporter.

Nature communications·2025
Same author

Large transient assemblies of Apaf1 constitute the apoptosome in cells.

Nature communications·2025

Related Experiment Video

Updated: Jan 22, 2026

Identification of Mouse and Human Antibody Repertoires by Next-Generation Sequencing
08:51

Identification of Mouse and Human Antibody Repertoires by Next-Generation Sequencing

Published on: March 15, 2019

13.0K

Deep generative modeling captures maturation-dependent pairing patterns in human antibodies.

Lea Brönnimann1,2, Thomas Lemmin1, Chiara Rodella1,2

  • 1Institute of Biochemistry and Molecular Medicine (IBMM), University of Bern, 3012 Bern, Switzerland.

Iscience
|January 21, 2026
PubMed
Summary

We developed a deep learning framework to predict antibody light chains from heavy chains. This method generates plausible antibody pairs, aiding immune repertoire analysis and therapeutic antibody design.

Keywords:
Artificial intelligenceImmunologyStructural biology

More Related Videos

Quantification of Antibody-dependent Enhancement of the Zika Virus in Primary Human Cells
07:53

Quantification of Antibody-dependent Enhancement of the Zika Virus in Primary Human Cells

Published on: January 18, 2019

9.3K
Generation of Human Adipose Stem Cells through Dedifferentiation of Mature Adipocytes in Ceiling Cultures
07:35

Generation of Human Adipose Stem Cells through Dedifferentiation of Mature Adipocytes in Ceiling Cultures

Published on: March 7, 2015

12.1K

Related Experiment Videos

Last Updated: Jan 22, 2026

Identification of Mouse and Human Antibody Repertoires by Next-Generation Sequencing
08:51

Identification of Mouse and Human Antibody Repertoires by Next-Generation Sequencing

Published on: March 15, 2019

13.0K
Quantification of Antibody-dependent Enhancement of the Zika Virus in Primary Human Cells
07:53

Quantification of Antibody-dependent Enhancement of the Zika Virus in Primary Human Cells

Published on: January 18, 2019

9.3K
Generation of Human Adipose Stem Cells through Dedifferentiation of Mature Adipocytes in Ceiling Cultures
07:35

Generation of Human Adipose Stem Cells through Dedifferentiation of Mature Adipocytes in Ceiling Cultures

Published on: March 7, 2015

12.1K

Area of Science:

  • Immunology
  • Bioinformatics
  • Computational Biology

Background:

  • Antibody heavy-light chain pairing is crucial for immune system function and therapeutic antibody development.
  • Existing sequence databases often lack essential paired antibody chain information.
  • Accurate pairing data is vital for understanding immune repertoire architecture.

Purpose of the Study:

  • To develop a computational framework for predicting antibody light chains based on heavy chain sequences.
  • To address the gap in paired antibody chain data within sequence databases.
  • To generate plausible antibody pairs for improved repertoire analysis and therapeutic design.

Main Methods:

  • A two-stage deep learning framework was employed.
  • Transformer-based language models were pre-trained on unpaired antibody sequences.
  • A sequence-to-sequence model was utilized to generate light chains from heavy chain inputs.

Main Results:

  • Generated antibody sequences demonstrated high germline identity and improved structural quality.
  • The model achieved moderate recovery of native light chains.
  • Analysis revealed maturation-dependent selection in heavy chains from memory B cells.
  • Generated kappa light chains showed a trimodal similarity distribution, suggesting distinct pairing modes.

Conclusions:

  • Sequence-to-sequence modeling can effectively uncover inter-chain dependencies in antibodies.
  • The developed framework can generate plausible antibody pairs, enhancing computational repertoire analysis.
  • This approach provides a foundation for designing novel therapeutic antibodies with specific pairing characteristics.