Data Mining and Computational Modeling of High-Throughput Screening Datasets

Abstract

We are now seeing the benefit of investments made over the last decade in high-throughput screening (HTS) that is resulting in large structure activity datasets entering public and open databases such as ChEMBL and PubChem. The growth of academic HTS screening centers and the increasing move to academia for early stage drug discovery suggests a great need for the informatics tools and methods to mine such data and learn from it. Collaborative Drug Discovery, Inc. (CDD) has developed a number of tools for storing, mining, securely and selectively sharing, as well as learning from such HTS data. We present a new web based data mining and visualization module directly within the CDD Vault platform for high-throughput drug discovery data that makes use of a novel technology stack following modern reactive design principles. We also describe CDD Models within the CDD Vault platform that enables researchers to share models, share predictions from models, and create models from distributed, heterogeneous data. Our system is built on top of the Collaborative Drug Discovery Vault Activity and Registration data repository ecosystem which allows users to manipulate and visualize thousands of molecules in real time. This can be performed in any browser on any platform. In this chapter we present examples of its use with public datasets in CDD Vault. Such approaches can complement other cheminformatics tools, whether open source or commercial, in providing approaches for data mining and modeling of HTS data.

Selecting machine-learning scoring functions for structure-based virtual screening

“…On the other hand, the number of HTS datasets has strongly increased in recent years [32,40]. If available for the target of interest, such dataset constitutes an interesting alternative as a test set (containing true instead of assumed inactives, better characterising the chemical diversity of molecules or representing by definition a realistic active-inactive proportion).…”

Extremely Low Hit Rate in a Diverse Chemical Drug Screen Targeting Mycobacterium abscessus

“…From a total of seven hits, only one showed a MIC below 10 M and moderate cytotoxicity. Five of the seven M. abscessus hits were either positive in our M. tuberculosis REMA or had been described previously as hits in other M. tuberculosis screening campaigns (16). Upon testing of M. abscessus hits against M. tuberculosis, we identified only one substance (number 2) which seems to possess selective activity against M. abscessus ( Table 1).…”

“…Hit compound 2, a carbamothioate, and hit compound 3, a thiourea, had MICs of 22 M and 23.7 M, respectively. IC 50 cytotoxicity values were Ͼ100 M. A database search revealed that both substances display known antituberculous activity (Collaborative Drug Discovery Vault) (16). This also holds true for the imidazo-pyridazine compounds 4 to 6, all which displayed an SI of Ͻ1 due to pronounced cytotoxicity.…”

A Web Repository System for Data Mining in Drug Discovery

“…Datasets" that is done by Ekins [18], "Web-based Drug Repurposing Tools" that is written be Sam E and Athri P [19], "DRUG Discovery Using Data Mining" that is provided by Charanpreet Kaur and Shweta Bhardwaj [20], and "Using Genetic Algorithms for Data Mining Optimization in an Educational Web-Based System" that is written by Behrouz Minaei-Bidgoli et al [21]. Some of these papers may provide part of the Data Mining functionalities we present in our work, but none are equipped with a strong backend database as we offer in our work.…”