Optimized Virtual Screening Workflow. Towards Target-Based Polynomial Scoring Functions for HIV-1 Protease

Val Oliveira Pintro, Walter F. Azevedo
  • Combinatorial Chemistry & High Throughput Screening, November 2017, Bentham Science Publishers
  • DOI: 10.2174/1386207320666171121110019

Development of a new scoring function to predict binding affinity for HIV-1 protease

What is it about?

We used the program SAnDReS to develop HIV-1 protease targeted scoring functions for prediction of binding affinity. These scoring functions were developed using machine learning methods implemented in the program SAnDReS (www.sandres.net).

Why is it important?

This is the first time that a machine-learning model was developed to predict binding affinity for HIV-1 protease using the experimental information from an ensemble of crystallographic structures available for complexes for which binding affinity data is available.


Walter de Azevedo Jr.

The methodology described in this paper makes available to the scientific community a reliable computational model to predict binding affinity for the HIV-1 protease. Also, the open source tools applied to create the machine-learning model for the HIV-1 protease can be applied to other protein targets.

Read Publication


The following have contributed to this page: Walter de Azevedo Jr.