TransportTP: A Tool for Predicting Membrane Transporters
Author Information
Author(s): Li Haiquan, Benedito Vagner A, Udvardi Michael K, Zhao Patrick Xuechun
Primary Institution: The Samuel Roberts Noble Foundation, Inc
Hypothesis
Can a two-phase classification approach improve the prediction and characterization of membrane transporters?
Conclusion
TransportTP is the most effective tool for eukaryotic transporter characterization to date.
Supporting Evidence
- TransportTP achieved an average recall of 81.8% and precision of 81.8% in cross-validation tests.
- TransportTP outperformed traditional methods like BLAST and HMM in transporter prediction.
- TransportTP's performance was validated using multiple model and non-model organisms.
Takeaway
TransportTP is a computer program that helps scientists find and understand proteins that move substances in and out of cells, making it easier to study them.
Methodology
TransportTP uses a two-phase classification approach combining homology-based methods and machine learning to predict transporters.
Potential Biases
Potential for false positives due to similarities with non-transporter proteins.
Limitations
The performance may vary based on the evolutionary distance between training and testing organisms.
Participant Demographics
Included model organisms like yeast and non-model organisms for testing.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website