Predicting HIV-1 Coreceptor Usage Using Structural Descriptors
Author Information
Author(s): Sander Oliver, Sing Tobias, Sommer Ingolf, Low Andrew J, Cheung Peter K, Harrigan P. Richard, Lengauer Thomas, Domingues Francisco S
Primary Institution: Max-Planck-Institute for Informatics, Saarbrücken, Germany
Hypothesis
Using structural information on the V3 loop in combination with sequence features improves prediction of HIV-1 coreceptor usage.
Conclusion
The proposed method significantly improves the prediction of HIV-1 coreceptor usage compared to traditional sequence-based methods.
Supporting Evidence
- The proposed method achieved a sensitivity of 0.80 when combined with sequence-based representations.
- Structural descriptors improved predictive performance over traditional methods.
- The study analyzed 514 distinct V3 sequences to assess coreceptor usage.
- Significant improvements in sensitivity were observed with a p-value of 0.0059.
- The method is robust against sequence variants containing insertions and deletions.
Takeaway
Scientists found a better way to predict how HIV enters cells by looking at the shape of a part of the virus, which helps in treating the disease.
Methodology
The study used structural descriptors derived from the V3 loop of the HIV-1 gp120 protein and compared their predictive performance against traditional sequence-based methods.
Potential Biases
Potential bias due to the reliance on specific sequence data and structural models that may not represent all viral variants.
Limitations
The method assumes a fixed backbone structure and does not account for side chain interactions within the V3 loop.
Participant Demographics
The dataset included 514 distinct V3 sequences from HIV-1 infected patients.
Statistical Information
P-Value
0.0059
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website