HIV-1 coreceptor usage prediction without multiple alignments: an application of string kernels
2008

Predicting HIV-1 Coreceptor Usage with String Kernels

Sample size: 1425 publication Evidence: high

Author Information

Author(s): Sébastien Boisvert, Mario Marchand, François Laviolette, Jacques Corbeil

Primary Institution: Centre de recherche du centre hospitalier de l'Université Laval

Hypothesis

Can support vector machines with string kernels effectively predict HIV-1 coreceptor usage without multiple alignments?

Conclusion

The SVM with the distant segments kernel is currently the best method for predicting HIV-1 coreceptor usage.

Supporting Evidence

  • The SVM with the distant segments kernel achieved an accuracy of 96.35% for CCR5 usage.
  • The SVM with the distant segments kernel achieved an accuracy of 94.80% for CXCR4 usage.
  • The SVM with the distant segments kernel achieved an accuracy of 95.15% for CCR5 and CXCR4 usage.

Takeaway

This study shows a new way to predict how HIV-1 uses different receptors to enter cells, which can help in treating the virus.

Methodology

The study used support vector machines with string kernels to classify HIV-1 coreceptor usage based on the V3 loop of the protein envelope sequence.

Potential Biases

Potential biases may arise from the dataset used, which could affect the generalizability of the findings.

Limitations

The study did not account for all potential biological factors influencing coreceptor usage.

Participant Demographics

The study analyzed HIV-1 sequences from various subtypes, primarily from the Los Alamos National Laboratory HIV Databases.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1742-4690-5-110

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication