Dr. Walter F. de Azevedo, Jr. has been ranked among the most influential researchers in the world  (Fields: Biophysics, Biochemistry, and Molecular Biology) according to a database created by Journal Plos Biology (see news here). The application of the same set of metrics recognized the influence of Dr. Azevedo's work in the following years (Baas et al., 2021; Ioannidis, 2022; Ioannidis, 2023). Regarding scientific impact (Peterson, 2005), Hirsch says that for physicists, a value for h index of 45 or higher could mean membership in the National Academy of Sciences of the USA. Dr. Azevedo is a physicist with an h-index of 50 in Scopus.  

His research interests are interdisciplinary, with three main emphases: artificial intelligence, complex systems, and computational systems biology. Dr. Azevedo developed several free software to explore the concept of Scoring Function Space. He has over 200 scientific publications about protein structures, computer models of complex systems, and simulations of protein systems. 

2021 data-update   2022 data-update   2023 data-update 

Updated on June 03, 2024. 

Exploring the Scoring Function Space [>>]

    A view of the scoring function space as a way to develop a computational model to predict ligand-binding affinity (Bitencourt-Ferreira et al., 2024).

We envisage protein-ligand interaction as a result of the relation between the protein space (Smith, 1970Hou et al., 2005) and the chemical space (Bohacek et al., 1996Dobson, 2004Kirkpatrick & Ellis, 2004Lipinski & Hopkins, 2004Shoichet, 2004Stockwell, 2004), and we propose to approach these sets as a complex system, where the application of artificial intelligence could contribute to understanding the structural basis for the specificity of ligands for proteins. Such approaches can create novel scoring functions to predict binding affinity with superior predictive power compared with classical scoring functions (also known as universal scoring functions) (Ross et al., 2013; Bitencourt-Ferreira et al., 2024; de Azevedo Jr. et al., 2024) available in docking programs (de Azevedo Jr., 2021). We propose to use the abstraction of a mathematical space composed of infinite computational models to predict ligand-binding affinity, named scoring function space (SFS) (Ross et al., 2013; Heck et al., 2017Bitencourt-Ferreira & de Azevedo Jr., 2019; Veríssimo et al., 2022; Bitencourt-Ferreira et al., 2024; de Azevedo Jr. et al., 2024).


With the development of the SFS concept, we expect to merge the holistic view of systems biology with machine-learning methods to contribute to drug discovery projects (Bitencourt-Ferreira et al., 2024; de Azevedo Jr. et al., 2024). By the use of supervised machine learning techniques, we can explore this SFS (Bitencourt-Ferreira & de Azevedo Jr., 2019da Silva et al., 2020de Azevedo Jr. et al., 2024) to build a computational model targeted to a specific protein system (targeted-scoring function) (Seifert, 2009). For instance, we created targeted-scoring functions for cyclin-dependent kinases (EC 2.7.11.22) (de Ávila et al., 2017; Levin et al., 2018da Silva et al., 2020de Azevedo Jr. et al., 2024) and HIV-1 protease (EC 3.4.23.16) (Pintro & de Azevedo, 2017).  

We developed the programs SAnDReS (Xavier et al., 2016; Bitencourt-Ferreira & de Azevedo Jr., 2019de Azevedo Jr. et al., 2024), SFSXplorer, and Taba (da Silva et al., 2020; Bitencourt-Ferreira et al., 2021) to generate computational models to predict ligand-binding affinity. These programs are integrated computational tools to explore the SFS (Bitencourt-Ferreira et al., 2024).