Data-driven discovery

Rajan K (ed) (2013) Informatics for materials science and engineering data-driven discovery for accelerated experimentation and applications. Elsevier, New York Hudson BD, Hyde RM, Rahr E, Wood J, Osman J (1996) Parameter based methods for compound selection from chemical databases. Quant Struct-Act Relat 15 285-289 Holliday JD, Willett P (1996) Definitions of dissimilarity for dissimilarity-based compound selection. J Biomolec Screen 1 145-151... 

The paradigm shift from critical activities from later drug development to earlier discovery phases some years ago has effectively led to a change in lead optimization and added a new dimension of complexity, while it is envisioned that from a multidimensional, data-driven process more suitable candidates in accord with the therapeutic target product profiles may emerge for the treatment of currently unmet medical needs. 

Wild, H. and Hinzen, B. (2006) Improving the hit-to-lead process data-driven assessment of drug-like and lead-like screening hits. Drug Discovery Today, 11, 175-180. 

Since modern drug discovery is mainly a data-driven process and chemoinformatics is at the center of data integration and utilization, it is natural that majority of library design tools are chemoinformatics tools. Therefore, a deep understanding of chemoinformatics is necessary for taking full advantage of library technologies. 

Data-driven or systems approaches have to rely on discovery science and hypothesis-driven research in individual fields of the life sciences and are complementary to them. Without knowledge gained with reductionist approaches, interpretation of data with the systems approach is not feasible. 

In hypothesis-driven discovery programs, often a list of candidate markers are compiled using the current knowledge base. The list of candidates may be compiled with advanced bioinformatics data mining tools (i.e., electronic interpretation of scientific papers), which could generate candidates that represent members of a probable pathway, either as expressed genes, or proteins, or the affected metabolites. 

Statistical HTS data analysis and discovery cheminformatics are integral components of an HTS-driven lead discovery process. Various statistical and visualization techniques are routinely... 

Accelerated approaches to data mining strategies19 involve sophisticated computer-driven data processing20 such as the X-hitting algorithm21 and cluster analyses.22 Each facilitates the rapid identification of active compounds while providing a measure of chemical diversity and novelty, thus reducing the discovery timeline. 

---