A data warehouse as an infrastructure to mine molecular descriptors for virtual screening

Abstract

In a Rational Drug Design (RDD) one important step is the receptor-ligand interaction evaluation through molecular docking simulations. How it is a way impossible to test all available compounds for a target receptor, there is a need to select the most promising. One possible approach for such selection is to consider characteristics like a set of molecular properties called molecular descriptors. Aiming at describing these characteristics, we introduce a Data Warehouse (DW) model that integrates molecular descriptors from different public databases of compounds, as well as relates them with Virtual Screening (VS) experiments data. With the proposed DW we are able to produce proper data sets for classification mining experiments. We performed a case study with a VS considering as receptor the HIV-1 Protease receptor and 76 compounds. The data sets produced from our DW are composed by 7 molecular descriptors as the predictive attributes, and as a target attribute the discretized Free Energy of Binding (FEB) value between the ligands and the target receptor. By performing C4.5 algorithm over the generated data sets, we got decision-trees models that indicates which molecular descriptors and their respective values are relevant to influence on good FEB results.

An Infrastructure to Mine Molecular Descriptors for Ligand Selection on Virtual Screening

“…From the diverse ligand databases we collect the molecular descriptors for each ligand being used. All data related to virtual screening and ligand databases modules are properly processed and inserted into the DW we have developed [ 6 ]. The stored data can then be preprocessed so that we can start the mining experiments.…”

“…This information is stored under a unified schema to facilitate decision making and built in a way to satisfy a multidimensional structure [ 30 ], also called analytical model. We have introduced a DW schema able to integrate molecular descriptors from different public ligand databases as well as able to relate them with virtual screening experiments data [ 6 ]. The main idea of our DW is not only to provide a single source capable of storing as many molecular descriptors as the ones provided by different public ligand databases but also to provide historic records of virtual screening experiments using molecular descriptors.…”

“…The mentioned DW [ 6 ] contains 6 dimension tables to represent the subject we are modelling, around a central fact table, called virtual screening. This DW schema can provide data about characteristics appearing as relevant in a virtual screening experiment, shared in the dimension tables about the whole experiment; molecular descriptors; database from with structures were collected; proteins; ligands and atoms.…”

“…Initially we proposed data warehouse (DW) schemas which are able to integrate molecular descriptors from different databases and relate them with VS experimental data [6]. With this powerful infrastructure we are encouraged to propose a methodology that makes use of the stored data to perform mining experiments on molecular descriptors characteristics.…”

“…With the aim to contribute to a more effective ligand selection, in this work we focus on a new strategy to easily and efficiently describe important characteristics that indicate promising compounds to be investigated in VS experiments. Initially we proposed data warehouse (DW) schemas which are able to integrate molecular descriptors from different databases and relate them with VS experimental data [ 6 ]. With this powerful infrastructure we are encouraged to propose a methodology that makes use of the stored data to perform mining experiments on molecular descriptors characteristics.…”