Computational Medicinal Chemistry – A Primer and Call for Contributions

Computational medicinal chemistry is a wide field covering many different methods including mainstays in the computational chemistry scene such as (3D-)QSAR or ligand docking as well as more recent or new developments including, among others, ADMET models, SAR visualization techniques, or multi-parameter optimization methods. In addition, compound data mining is increasingly being recognized as a valuable source of information for medicinal chemistry projects.

Like Comment

QSAR and docking have long been used in medicinal chemistry to support compound optimization or identify new active molecules. Computations making direct contributions to compound identification or optimization are typically classified as prospective applications. In addition, 3D-QSAR and docking are frequently used in the literature to rationalize observed SARs or predict compound binding modes. Such modeling is retrospective in nature. For medicinal chemistry publications, retrospective computational studies often provide visually appealing components (add-ons) that might well increase the attractiveness of a manuscript. However, such retrospective modeling exercises are usually hypothetical in nature and have only little impact on practical medicinal chemistry.

The true impact of computational methods on the practice of medicinal chemistry is often difficult to communicate and assess in scientific publications. A major reason is that most projects in pharmaceutical R&D are proprietary and cannot be disclosed to the extent required to present a paper conforming to general publication, disclosure, and data deposition standards. Publications are often only possible once a project (area) has been terminated. Furthermore, a publication culture is, unfortunately, still not supported by all pharmaceutical companies. Moreover, many investigators intimately involved in project team work and routine tasks (e.g., computational compound screening, maintenance of computational infrastructures etc.) do not find the time (or are not motivated) to invest in scientific publications. Although novel computational concepts and exemplary applications are frequently published, it is evident that the practice of computational medicinal chemistry is not necessarily reflected in the current scientific literature.

MedChemNet should also provide a forum for exchanging views on practical issues related to computational medicinal chemistry, drug design, or compound data mining and for interactively assessing the state-of-the-art in this area, without the need to disclose proprietary information. It would be very nice and informative to see contributions from computational scientists in pharma settings that provide some insights into their day-to-day efforts, sharing experience values of how computational approaches impact medicinal chemistry and drug discovery projects in their specific environments. For example, which computational methods work for us, which ones do not? Which of the computational approaches available to us make a true impact on chemistry/discovery projects and how? Which of the problems we face are notoriously difficult to address using currently available software tools? Where do we see the need to engage in the development of our own methods? These and related questions could be addressed in individual contributions, which should best have the format of brief communications (with perhaps 500-1000 words). I envision that one would collect such posts as templates and ultimately publish them jointly in a review-type article in Future Medicinal Chemistry, similar in design and spirit to an earlier report that has made an important contribution to the diversity analysis field [1]. This would further increase the visibility and weight of such off the beaten path communications and their weight. Another interesting format for following up on relevant posts and further personalizing them would be provided by an “ask the experts”-type contribution [2], as introduced by Future Medicinal Chemistry For the practice of computational medicinal chemistry and drug design, exchanging and collecting reports and views on the type of questions raised above should represent an informative account and important step forward.

[1] Martin YC et al. Diverse viewpoints on computational aspects of molecular diversity. J Comb Chem 3(3), 232-250, 2001.

[2] Bajorath J et al. Ask the experts: focus on computational chemistry. Future Med Chem 3(8), 909-921, 2011.

Go to the profile of Jürgen Bajorath

Jürgen Bajorath

Professor , University of Bonn

For further details, see:

No comments yet.