I recently set out to drive from Columbus, OH, to Minneapolis, MN for some family festivities. Along the way, I wanted to work in a nice stop for lunch, specifically, a small town, family-run restaurant with a unique feel and, of course, great pie.
In these scenarios, modern technology is incredible. It provides unprecedented access to relevant information that can help with this task in real-time. I scanned through Waze, Yelp, TripAdvisor®, and Google search results, which yielded an overwhelming amount of information to help me choose. And choose I did: A little spot in Illinois. But as I dug deeper into my research and my meal, I discovered that the absolute best spot, the place with the most coveted apple pie in Illinois, a bona fide national treasure, was right across the street.
I wondered: Wouldn’t it be amazing to have a tool that cut through the deluge of information and just said, “This is exactly what you’re looking for.”
Narrowing down the synthetic routes …
This scenario plays out often in the lives of synthetic chemists. Retrosynthesis planning can take hundreds of hours per target compound, so chemists learn early on that it’s important to first assess all prior experiments and data. This process would not be possible without tools like SciFindern - already used by the world’s synthetic chemists to quickly and easily access the most complete collection of synthetic information. Yet sifting through it all to find and compile the perfect synthetic plan remains insurmountable. For example, we have almost 120 million reactions extracted from patent and journal literature in our content collection. No one can manually process this much information, let alone connect the dots between the documents.
What the field needs is a technology solution that can aid the chemist by shouldering the herculean task of assembling, analyzing and prioritizing all the relevant synthetic routes extracted from the patent and journal literature extracted for a molecule, allowing them free up their time to get to the lab and move their research forward. This is not a new idea, but it’s increasingly urgent given the rise of big data and the need for novel compounds in the pharmaceutical and specialty chemicals industries. So, what’s happening in this space?
Retrosynthesis software: 50 years in the making
Computer-aided synthetic design (CASD) has, in principle, presented solutions to many problems in retrosynthesis planning for more than 50 years. However, as the apple pie analogy illustrates, too often these tools can get you close, but not close enough. This may not be the case anymore. The development and deployment of new resources for retrosynthesis are reaching exciting new levels.
On our end, CAS is developing innovative tools and services to address the limitations of currently available retrosynthetic planning solutions. Our expertise and experience in the field of organic synthesis has allowed us to approach this in a balanced way, using novel algorithms in conjunction with highly comprehensive and accurate reaction content.
Significant advances have been made through our deeply researched and designed solution, SciFindern, created with the ability to piece together portions of the synthetic routes described in disparate journal articles and patents, uncovering new routes that were previously hard (if not impossible) to find. By mining the published literature, we can now provide users with an advanced retrosynthesis planning capability that can uncover, organize, and prioritize full retrosynthetic routes to known compounds. It draws on the best-known collection of evidence-based rules to prioritize and score the available routes to present you with the most relevant option. You can seamlessly jump and investigate other synthetic routes with just a few mouse clicks at your desk, lab bench or even at home.
From augmenting to reinventing synthetic routes
CAS has a transformative vision for what can be done for synthetic chemists. We’re working towards a very near-term future in which our platform can predict novel retrosynthetic routes for target molecules that have no published experimental evidence. We started with finding hidden or obscured synthetic routes to known compounds, and in the coming months we’ll provide the means to predict novel routes to novel compounds that only exist in the discovery chemist’s mind. This will be transformative, building on a platform focused on pure productivity and efficiency to a solution that inspires creativity and innovation; a platform that can drive that next “aha” moment. Success here would represent a massive advance for many fields and could lead to a reduction in the overall time-to-market for a number of products, including pharmaceuticals and specialty chemicals.
In this day and age, we have so many options – whether you’re looking for optimum synthetic routes or the best apple pie in the Midwest. To determine where to eat, I needed to wade through all the insights and reviews and locations of various lunch options. What has been missing is a retrosynthesis planning solution that can seek out the best pick on our behalf. That’s the stage we’re entering now in synthetic chemistry.