Integrating linear optimization with structural modeling to increase HIV neutralization breadth

• Published in: PLoS computational biology 2018-02, Vol.14 (2), p.e1005999-e1005999
• Main Authors: Sevy, Alexander M, Panda, Swetasudha, Crowe, Jr, James E, Meiler, Jens, Vorobeychik, Yevgeniy
• Format: Article
• Language: English
• Subjects: Algorithms; Amino Acid Motifs; Amino acid sequence; Amino acid sequencing; Antibodies; Antibodies, Neutralizing - immunology; Artificial intelligence; Binding; Bioinformatics; Biology and Life Sciences; Computation; Computational Biology; Computer and Information Sciences; Computer applications; Computer engineering; Computer science; Design; Electrical engineering; Engineering and Technology; Engineers; Epitopes - immunology; Grants; HIV; HIV Antibodies - immunology; HIV Infections - immunology; HIV-1; Human immunodeficiency virus; Humans; Immunoglobulins; Integer programming; Learning algorithms; Linear Models; Linear programming; Machine Learning; Mathematical models; Medicine and Health Sciences; Methods; Neutralization; Optimization; Optimization theory; Proteins; Publishing; Regression Analysis; Research and Analysis Methods; Sampling methods; Software; Software packages; Structural equation modeling; Supervision; Support Vector Machine
• ISSN: 1553-7358; 1553-734X; 1553-7358
• DOI: 10.1371/journal.pcbi.1005999

Computational protein design has been successful in modeling fixed backbone proteins in a single conformation. However, when modeling large ensembles of flexible proteins, current methods in protein design have been insufficient. Large barriers in the energy landscape are difficult to traverse while redesigning a protein sequence, and as a result current design methods only sample a fraction of available sequence space. We propose a new computational approach that combines traditional structure-based modeling using the Rosetta software suite with machine learning and integer linear programming to overcome limitations in the Rosetta sampling methods. We demonstrate the effectiveness of this method, which we call BROAD, by benchmarking the performance on increasing predicted breadth of anti-HIV antibodies. We use this novel method to increase predicted breadth of naturally-occurring antibody VRC23 against a panel of 180 divergent HIV viral strains and achieve 100% predicted binding against the panel. In addition, we compare the performance of this method to state-of-the-art multistate design in Rosetta and show that we can outperform the existing method significantly. We further demonstrate that sequences recovered by this method recover known binding motifs of broadly neutralizing anti-HIV antibodies. Finally, our approach is general and can be extended easily to other protein systems. Although our modeled antibodies were not tested in vitro, we predict that these variants would have greatly increased breadth compared to the wild-type antibody.