Jarl Wikberg - ICPB2016

"Pharmaceutical bioinformatics"

Is there a distinct field - Pharmaceutical Bioinformatics? We argue that there is and that there is a need for it. Bioinformatics is the field of understanding biological data. It originated in the 1960s, essentially from the analysis of sequence data and has since then evolved into many important fields, including genome, transcriptome, proteome and metabolome analysis, structural bioinformatics, network and system biology, and others, areas which are indispensable in understanding biological function, diseases, evolution, etc. Cheminformatics on the other hand is an important area in chemistry and pharmacy which origin is the indexing of chemical data that commenced even long before the advent of computers, namely at the end of the 1800s century with the Chemical Abstracts (later CAS) and then into the field of cheminformatics which is devoted to the study of chemistry and chemical compounds with computers. Drugs are cornerstone in medicine, but there is still a large unmet need for new molecular entities (NMEs) for treatment of major diseases, rare diseases, neglected diseases, to combat drug resistance, etc. From 1930 up until now FDA has approved about 1,400 NMEs, directed to less than totally 400 targets, with the number of novel NMEs registered each year varying between about 20-30. Currently developed drugs cover only about 3% of the druggable genome (see abstract by Tudor Oprea at this conference). Thus, covering all the opportunities will take quite some time, considering the rising development costs and harsher governmental budget constraints, making it more and more difficult to cover the development costs, in particular for rare and neglected diseases. With the current rate this may actually never happen. Our interest in the use of informatics to make drug discovery and development more efficient and less costly commenced with our discovery of proteochemometrics in 2002. In short proteochemometrics merges bioinformatics sequence information or structure information with cheminformatics descriptions of chemical entities. From this an ‘interaction space’ can be computed, which together with measured activity data can provide unified informatics models that generalize into the interaction of drugs with broad target groups, even entire genomes. Although such models should obviously have immense impact on understanding biology, and be highly useful to improve the development of drugs, quite to our surprise communicating our results with the biological, medical, pharmaceutical and chemical communities proved very difficult. The need for education came immediately clear, and to this end we formed the study subject ‘Pharmaceutical Bioinformatics’ at our faculty. Chemobioinformatics is now used among virtually the entire pharmaceutical industry in their development pipelines (‘predictive sciences’), predictive computations are mandatory in regulatory processes, and informatics is applied in virtually all areas of biomedicine. It is, however, our belief that the study of informatics in all areas of science related to drugs; their discovery and development; elucidating and understanding their function (‘pharmacology’); and promoting their safety and use in therapy, should have interdisciplinary forum for advancement, application, education and exchange. While names can to some extent be arbitrary, we propose that ‘Pharmaceutical Bioinformatics’ can make a platform for this. It
is our hope that many will contribute to this exciting field.