By: Aiden Aceves
Talk to any group of people working in the field, and it becomes immediately clear that there is no one definition of “cheminformatics.” I will start with mine: cheminformatics is the sum of tools and algorithms for managing and manipulating chemical data, including structures, physical properties, and even patterns of activity.
From big pharma’s gargantuan digital inventories to academic and biotech efforts to decode structural activity, there is little doubt that cheminformatics as a field has shaped drug discovery since its inception. In the 1940s, the chemical processing potential of the computer was quickly realized by scientists who used the newly invented devices to model molecular rotors. The decades that followed saw tremendous advances in cheminformatics, including the forerunners of substructure matching algorithms, and the creation of the database technology and standards for the storing of chemical data. (A more complete history can be read elsewhere.)
While the field has existed for more than half of a century, the “cheminformatics” name itself is relatively new and was crafted to build upon the surging scientific and public popularity of bioinformatics in the 1990s. Although bioinformatics also traces its origins to the mid-20th century, the burgeoning genetic data made possible by the technologies of the 1990s made the field one of the fastest growing areas of biomedical research. By many metrics (jobs, funding, publications), the field of bioinformatics has outpaced cheminformatics over the last 15 years, leading some to even consider cheminformatics a sub-discipline of bioinformatics. Regardless of the organizational hierarchy of the fields, there is no doubt that there is a large degree of overlap between the fields, and in that overlap lies potential for advancement and growth.
Just as bioinformatics exists to process and make sense of the incredible volumes of data generated by the biomedical sciences, cheminformatics can be used to interface the world of small molecules with this wealth of data and make meaningful connections. As part of my work at Evince Biosciences, I have worked to develop strategies for characterizing small molecules and the ways in which they relate to the “big data” of bioinformatics. By tapping into this data oasis, we have been able to build panels of simulations that provide a canonical fingerprint describing chemical behavior across entire pathways, from target to the whole organism level.
As a biotech startup, being able to parse public data using the techniques of bioinformatics and cheminformatics has allowed us to step out from the shadow of bioinformatics and develop a platform where the two disciplines come together to create a comprehensive digital model of drug activity. This approach is being developed as the next generation of drug discovery.