Functional Representation of Enzymes by Specific Peptides
Author Information
Author(s): Kunik Vered, Meroz Yasmine, Solan Zach, Sandbank Ben, Weingart Uri, Ruppin Eytan, Horn David
Primary Institution: Tel Aviv University
Hypothesis
Can specific peptides extracted from enzyme sequences accurately predict enzyme functions?
Conclusion
The study demonstrates that specific peptides can classify 93% of enzymes, significantly improving upon previous methods.
Supporting Evidence
- Specific peptides cover 65% of all active sites in enzymes.
- The method classifies enzymes with 93% accuracy.
- SPs provide better coverage than traditional ProSite motifs.
- SPs are found in active-site neighboring 3-D pockets.
- SPs can classify enzymes even when sequence similarity fails.
Takeaway
Scientists found short pieces of proteins that help tell what the proteins do, covering most known enzymes.
Methodology
An unsupervised motif extraction algorithm (MEX) was applied to enzyme sequences to identify specific peptides.
Potential Biases
There is a risk of bias due to the presence of enzymes in the test set that are similar to those in the training set.
Limitations
The study may be biased due to high sequence similarity among enzymes in the training and test sets.
Statistical Information
P-Value
0
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website