Predicting Protein Kinase Substrates Using Amino Acid Sequences
Author Information
Author(s): Sobolev Boris, Filimonov Dmitry, Lagunin Alexey, Zakharov Alexey, Koborova Olga, Kel Alexander, Poroikov Vladimir
Primary Institution: Institute of Biomedical Chemistry of the Russian Academy of Medical Sciences
Hypothesis
Can the PAAS method accurately predict protein kinase substrates to enrich signaling pathway networks?
Conclusion
The PAAS method effectively predicts protein kinase substrates, enhancing the understanding of signaling pathways.
Supporting Evidence
- The PAAS method achieved reasonable accuracy in predicting kinase-specificity classes.
- Predictions were validated against known interactions in the TRANSPATH® database.
- Newly predicted interactions helped reconstruct potential signaling pathways.
Takeaway
This study shows a way to guess which proteins are affected by certain enzymes, helping scientists understand how cells communicate.
Methodology
The PAAS method was validated using leave-one-out cross-validation and multiple splitting of datasets to predict kinase-substrate interactions.
Potential Biases
Potential bias due to reliance on existing databases for training data.
Limitations
The method may not accurately predict interactions for highly variable kinase substrate classes.
Participant Demographics
The training set included 971 mammalian proteins, primarily human, and other vertebrates, fungi, viruses, and insects.
Statistical Information
P-Value
0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website