Figures
Predicting HIV-1 neutralization by antibodies using machine learning
Broadly neutralizing antibodies against HIV-1 are currently investigated in clinical trials as a new treatment option. In order to select an effective antibody therapy, the neutralization sensitivity of the patient's viral strains towards the antibodies must be ensured. Since neutralization assays are too time-consuming and expensive, they are not suitable for routine clinical practice. Applying machine learning on existing neutralization assay data, Hake and Pfeifer developed a model for accurately predicting the neutralization sensitivity of HIV-1 towards antibodies based on the viral envelope sequence. The illustration sketches how the classifier discriminates between susceptible and resistant samples.
Image Credit: Anna Hake
Citation: (2017) PLoS Computational Biology Issue Image | Vol. 13(11) November 2017. PLoS Comput Biol 13(11): ev13.i11. https://doi.org/10.1371/image.pcbi.v13.i11
Published: November 30, 2017
Copyright: © 2017 Anna Hake/Max Planck Institute for Informatics. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Broadly neutralizing antibodies against HIV-1 are currently investigated in clinical trials as a new treatment option. In order to select an effective antibody therapy, the neutralization sensitivity of the patient's viral strains towards the antibodies must be ensured. Since neutralization assays are too time-consuming and expensive, they are not suitable for routine clinical practice. Applying machine learning on existing neutralization assay data, Hake and Pfeifer developed a model for accurately predicting the neutralization sensitivity of HIV-1 towards antibodies based on the viral envelope sequence. The illustration sketches how the classifier discriminates between susceptible and resistant samples.
Image Credit: Anna Hake