Related Experiment Video
Updated: Aug 5, 2026

Identification of Kinase-substrate Pairs Using High Throughput Screening
Published on: August 29, 2015
An AI-enabled structural atlas decodes kinase specificity across the human proteome
David R Vanderwall1, Edward L Huttlin1, Julian Mintseris1
1Department of Cell Biology, Harvard Medical School, Boston, MA, USA.
KinoPlex predicts which protein residues are likely phosphorylated and by which kinases, revealing new principles of kinase specificity. This computational framework maps phosphorylation potential and kinase recognition across the human proteome.
Area of Science:
- Biochemistry
- Proteomics
- Computational Biology
Background:
- Only a small fraction of human protein phosphorylation sites are experimentally validated and linked to specific kinases.
- Understanding kinase-substrate interactions is crucial for deciphering cellular signaling pathways.
Purpose of the Study:
- To develop KinoPlex, a computational framework for predicting phosphorylation potential and kinase specificity for all serine/threonine/tyrosine residues in the human proteome.
- To identify novel kinase-substrate interactions and uncover organizing principles of phosphorylation.
Main Methods:
- Integration of predicted protein structures (AlphaFold models) with kinase recognition motifs (position-specific scoring matrices).
- Application of positive-unlabeled transfer learning to identify structurally phospho-competent residues.
- Quantification of motif specificity to identify high-confidence kinase-substrate candidates.
Main Results:
- Identified approximately 567,000 structurally phospho-competent residues.
- Predicted around 250,000 high-confidence candidates with both sequence recognition potential and optimal structural presentation.
- Discovered the "sequence-structure selective coupling" phenomenon explaining kinase specificity based on motif structural accessibility.
Conclusions:
- KinoPlex provides a comprehensive atlas of potential phosphorylation sites and kinase specificities.
- The study reveals fundamental principles governing kinase substrate recognition and phosphorylation dynamics.
- Experimental validation in K562 cells supports the accuracy and utility of KinoPlex predictions.
Related Concept Videos
MAPK Signaling Cascades
Protein Kinases and Phosphatases
Protein kinases
Many proteins in the cell are regulated by phosphorylation, the addition of a phosphate group. A family of enzymes called kinases...
Protein Networks
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
Amplifying Signals via Enzymatic Cascade
The JAK-STAT Signaling Pathway
PI3K/mTOR/AKT Signaling Pathway

