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

Protein Kinases and Phosphatases02:54

Protein Kinases and Phosphatases

Proteins undergo chemical modifications that trigger changes in the charge, structure, and conformation of the proteins. Phosphorylation, acetylation, glycosylation, nitrosylation, ubiquitination, lipidation, methylation, and proteolysis are various protein modifications that regulate protein activity. Such modifications are usually enzyme-driven.
Protein kinases
Many proteins in the cell are regulated by phosphorylation, the addition of a phosphate group. A family of enzymes called kinases...
Protein Diffusion in the Membrane01:24

Protein Diffusion in the Membrane

Proteins show rotational as well as lateral diffusion across the membrane. The lateral diffusion of proteins was confirmed through the cell fusion experiment where mouse and human cells were fused, resulting in hybrid cells. When the human and mouse cells fused, the specific membrane proteins on human and mouse cells were marked with the red and green-fluorescent markers, respectively. Initially, the red and green fluorescence was located on the respective hemisphere of the cell. As time...
Protein Dynamics in Living Cells01:19

Protein Dynamics in Living Cells

Different fluorescence-based techniques are used to study the protein dynamics in living cells. These techniques include FRAP, FRET, and PET.
Fluorescent recovery after photobleaching (FRAP) is a fluorescent-protein-based detection technique used to quantify protein movement rates within the cell. This method exposes a small portion of the cell to an intense laser beam. The laser beam causes permanent photobleaching of the fluorophore-tagged proteins in the exposed region. As the bleached...

You might also read

Related Articles

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

Sort by
Same author

Contrastive diffusion model for exploring mathematical expressions from data.

Neural networks : the official journal of the International Neural Network Society·2026
Same author

Point Cloud Video Modeling With Progressive Prior Knowledge Guidance and Adaptive Neighboring Aggregation.

IEEE transactions on neural networks and learning systems·2026
Same author

Find Hidden Modality Divergence: Adversarial Aware Learning for Unsupervised Visible-Infrared Person Re-Identification.

IEEE transactions on neural networks and learning systems·2025
Same author

Protein Subcellular Localization Prediction Model Based on Graph Convolutional Network.

Interdisciplinary sciences, computational life sciences·2022
Same author

Refined Contact Map Prediction of Peptides Based on GCN and ResNet.

Frontiers in genetics·2022
Same author

Amino acid environment affinity model based on graph attention network.

Journal of bioinformatics and computational biology·2021

Related Experiment Video

Updated: May 26, 2026

Identification of Cyclin-dependent Kinase 1 Specific Phosphorylation Sites by an In Vitro Kinase Assay
12:26

Identification of Cyclin-dependent Kinase 1 Specific Phosphorylation Sites by an In Vitro Kinase Assay

Published on: May 3, 2018

KSDiffusion: conditional diffusion for kinase-specific phosphorylation site prediction under data-limited and

Shi Qiu1, Chunguo Wu1, Yuxiang Ma1

  • 1College of Computer Science and Technology, Jilin University, 2699 Qianjin Street, Chaoyang District, Changchun 130012, China.

Briefings in Bioinformatics
|May 25, 2026
PubMed
Summary

KSDiffusion enhances kinase-specific phosphorylation site prediction, especially for rare kinases, by combining protein language models with generative AI. This approach overcomes data limitations and improves accuracy in cellular signaling research.

Keywords:
ESM-2diffusion modelsfew-shot learningspecific kinase phosphorylation sites

More Related Videos

Identification of Kinase-substrate Pairs Using High Throughput Screening
11:13

Identification of Kinase-substrate Pairs Using High Throughput Screening

Published on: August 29, 2015

Related Experiment Videos

Last Updated: May 26, 2026

Identification of Cyclin-dependent Kinase 1 Specific Phosphorylation Sites by an In Vitro Kinase Assay
12:26

Identification of Cyclin-dependent Kinase 1 Specific Phosphorylation Sites by an In Vitro Kinase Assay

Published on: May 3, 2018

Identification of Kinase-substrate Pairs Using High Throughput Screening
11:13

Identification of Kinase-substrate Pairs Using High Throughput Screening

Published on: August 29, 2015

Area of Science:

  • Biochemistry and Molecular Biology
  • Computational Biology
  • Bioinformatics

Background:

  • Protein phosphorylation is crucial for cellular signaling, but identifying specific kinase-substrate sites is challenging due to limited and imbalanced experimental data.
  • Existing computational methods often rely on local sequence patterns, failing to capture broader contextual information needed for accurate prediction.
  • Data scarcity, particularly for underrepresented kinases, significantly hinders the performance of current predictive models.

Purpose of the Study:

  • To develop a novel computational framework, KSDiffusion, for accurate kinase-specific phosphorylation site prediction.
  • To address the challenges posed by data scarcity and imbalance in kinase-substrate datasets.
  • To improve the generalization performance of predictive models, especially for underrepresented kinases.

Main Methods:

  • KSDiffusion integrates a protein language model (ESM-2) for context-aware peptide representation extraction, incorporating evolutionary and structural information.
  • Supervised contrastive learning is employed to enhance kinase-specific discriminability within the learned embedding space.
  • A conditional diffusion model (KS-DiT) generates synthetic, biologically plausible representations to augment data for rare kinase groups.

Main Results:

  • KSDiffusion consistently outperformed existing baseline methods across various data regimes (low, medium, and large).
  • Significant improvements were observed for data-scarce kinase groups, with Area Under the Curve (AUC) gains up to approximately 15%.
  • The model maintained competitive performance even when ample training data were available, demonstrating its robustness.

Conclusions:

  • KSDiffusion offers a robust solution for kinase-specific phosphorylation site prediction, particularly effective in data-limited and imbalanced scenarios.
  • Conditional diffusion-based augmentation proves valuable and regime-dependent, enhancing predictions for underrepresented kinases.
  • The integration of protein language models with task-aware generative modeling represents a significant advancement for addressing realistic constraints in biological data analysis.