How much does drug discovery software cost?
How number of users affect drug discovery software costs The number of people who need access to the platform is…
What exactly is a medicinal chemist’s role today?
Firstly, let’s define a medicinal chemist’s role – to design and synthesise the most appropriate molecule for their project’s therapeutic objective as quickly and efficiently as possible.
The ultimate goal remains constant: identify lead molecules that are most likely to succeed through the drug development pipeline while avoiding costly late-stage failures.
However, the way that medicinal chemists approach this challenge is changing. Traditionally, using computational tools to model molecules and their properties was left to “computational” chemists. But in today’s landscape, we’re seeing that the lines have blurred. The role of a modern medicinal chemist incorporates multiple disciplines – part chemist, part data scientist, and part structural modeller. Catch our on-demand panel discussion ‘How to be great drug discovery chemist’ to hear from experts on the skills it takes to gain success in medicinal chemistry.
A word of reassurance: Software will not replace you
Let’s first offer some peace of mind. There are concerns around computational tools replacing human chemists’ jobs, especially at the rate at which technology advances.
However, we can’t underestimate the importance of a medicinal chemist – carrying a unique toolbox of experience, skills, and insight to make quick and informed decisions based on project context.
What software will do is help you do your job better and faster. It will enable you to make better use of your time and get more value from your experiments and data. Software brings speed and objectivity and enables you to work at a scale you cannot do alone.
How can software support your medicinal chemistry?
1. Gain new insights from complex data
Computers are much more adept at handling the volume and variety of information we see in drug discovery. They can provide powerful data visualisation and analyses to find and highlight patterns and trends that are not obvious in your raw data. This not only gives you new insights that advance your project but also helps you communicate your findings clearly with colleagues.
Structure-activity relationships (SAR) are essential patterns to identify in your molecules and data. Analyses, such as R-group decomposition, matched molecular pairs, and activity cliff detection, can quickly find SAR that guide your optimisation strategies. Additionally, by incorporating machine learning into your workflow, quantitative SAR (QSAR) models enable you to predict the most promising structural modifications to guide your next experiments.
Developing new drugs means balancing multiple parameters, including target activity, ADMET properties, selectivity, and safety. That’s far too many dimensions to manage in your head or put on a single graph. Software excels here for multi-parameter optimisation, balancing these parameters to identify the optimal compounds based on your specific criteria.
2. Thinking in 3D
Many medicinal chemists still work primarily in 2D. By ignoring the 3D structure of your molecules, you’re missing out on valuable information about how their spatial characteristics impact their binding interactions. To do this, you must use the right software to generate and visualise 3D structures.
3D information is valuable for both structure-based drug design (when protein structure is available) and ligand-based methods (when it isn’t).
In structure-based design, molecular docking predicts how compounds bind to targets and ranks their potential, incorporating hydrophobic interactions, hydrogen bonds, and ligand strain to help filter compound libraries. Even without target structural information, 3D analysis can align molecules in 3D to highlight their similarities and differences, helping identify new leads, enabling scaffold hopping, and guiding lead optimisation.
While you can leave this work to modellers, generating your own 3D hypothesis means you can get immediate feedback on your compounds and ideas, which can help you design better molecules faster.
3. Apply knowledge beyond your experience
Powerful cheminformatics and AI methods are no longer out of reach for medicinal chemists, and they offer huge potential to capture information from vast quantities of existing data and translate this into insights for your compounds and projects.
For example, matched series analysis identifies chemical substitutions likely to improve activity by examining series compounds that differ by just a single change from large databases. It allows you to extract more knowledge from existing data by linking structural modifications to changes in properties and applying these in the context of your series to propose new, relevant compounds.
For finding new ideas, generative chemistry methods enable you to explore vast chemical space at an unprecedented scale. They can suggest ideas that are likely to succeed based on your requirements. It does this while avoiding the cognitive biases that can limit human creativity, often revealing opportunities you might otherwise miss.
4. Avoid pitfalls with retrosynthetic analysis
It’s all well and good to identify a molecule that is predicted to sit favourably within your desired parameters, but it also needs to be synthetically feasible. Retrosynthetic analysis software helps you plan and evaluate synthetic routes, enabling you to rule out any compounds that just aren’t possible to synthesise or find new routes you may not have considered.
The importance of usable software
If you’re a medicinal chemist looking to incorporate software into your work, look for providers who prioritise usability. Even the most powerful tools are meaningless if they’re too complex to understand and use regularly.
The most valuable tools provide interactive experiences with quick feedback on ideas and queries. This enables you to iterate quickly to find the next optimal compound.
About the author
Matthew Segall, PhD
CEO and Company Director
Matt holds a Master’s in Computation from the University of Oxford and a PhD in Theoretical Physics from the University of Cambridge. He led teams developing predictive ADME models and advanced decision-support tools for drug discovery at Camitro (UK), ArQule Inc., and Inpharmatica. In 2006, he took charge of Inpharmatica’s ADME business, overseeing experimental services and the StarDrop software platform. After Inpharmatica’s acquisition, he became Senior Director of BioFocus DPI’s ADMET division and, in 2009, led a management buyout to establish Optibrium.
More about StarDrop
How do StarDrop and LiveDesign compare?
No glove fits every hand perfectly—and it’s the same with software tools. There’s no one-size-fits-all solution when it comes to…
How can I run a successful virtual screen?
Virtual screening is often powerful because it is fast (i.e., cheap) and can give new insights into your project much…
