00:00  Mark Meyers

Data has been stored in a way that only allowed for human points of access or interface. The emergence of machine learning, or ML, requires not human points of access, but machine points of access. And in order to achieve such machine access, it required data stakeholders to carefully consider the entire data infrastructure.

00:37  Jesse Harris

I feel like I say this every year, but 2022 was incredibly busy.

00:42  Sarah Srokosz

You can say that again. But now that this year is drawing to a close, it seems like a good time to reflect on how this year was shaped by the many disruptions of the years before and consider what is in store in the year to come.

00:55  Jesse Harris

Hi, I’m Jessie.

00:56  Sarah Srokosz

And I’m Sarah. We’re the hosts of the Analytical Wavelength. At ACD/Labs we get a unique perspective of what is happening in the world of chemistry.

01:06  Jesse Harris

Absolutely. We get to work with the best and the brightest in all levels of the chemical and pharmaceuticals industries. This gives us the opportunity to see what scientists all around the world are doing in the lab and how they use and manage their chemical data.

01:21  Sarah Srokosz

As the vice president of sales at ACD/Labs, Mark Myers It’s always following the trends to anticipate the needs and desires of our customers.

01:30  Jesse Harris

Today, we’re talking with him about what he sees as the most important takeaways from 2022 and what to expect from 2023. Let’s hear what he has to say.

01:40  Sarah Srokosz

Hello, Mark. Thank you so much for joining us on the podcast today. We’re going to start off with the question we ask all of our guests, and that is, what is your favorite chemical?

01:52  Mark Meyers

Well, that is an interesting question for a non-chemist. And my answer would be a chemical substance that I was once introduced to years ago by a colleague of mine. When I first entered the field of cheminformatics, and that compound is called a buckyball. Its shape is, you may or may not know is very interesting and consists of 60 carbon atoms, fused together in a soccer ball shape. Kind of timely given the World Cup going on. I recently reviewed an article online indicating that they actually thought that that these only existed in the lab and not in nature. But in 2010, astronomers actually found proof that they were indeed popping around in space, deep space, that is. And I’m sure we’re going to learn more about this mysterious compound in years to come.

02:43  Jesse Harris

Yeah, it is a really clever, interesting little thing, has lots of uses too, so it’s a very good choice. So let’s get into things. And you are of course, our VP of sales here. So we want to talk to you about some things related to the business side of the chemistry world. One important part of the sales process at ACD/Labs is understanding the workflows and business practices of our customers and deliver solutions that fit them.

Of course, there’s been a lot of disruptions in the last couple of years as everybody knows. So I was wondering how workflows and business practices have changed in the last couple of years from what you’ve seen?

03:21  Mark Meyers

It certainly is critical that we understand our customers workflows, but we also need to be intimately familiar with their data flows. Our sales organization is focused on the delivery of solutions that create value for our customers. And you can’t do any of this unless you invest time and understanding in customers workflows, processes, data flows and overall business and IT goals.

Each organization has its own unique underpinning from both an IT and business perspective, and these have to be understood to deliver these value-added solutions for them. Now, to answer your question, if I had to identify some specific changes in the industry dynamics related to workflow and business processes, it would be around the specific trends that have been accelerated due to the worldwide pandemic that we’ve all recently been through.

These trends include the automation of various processes and workflows in the lab that were previously performed manually. And again, all due to the direct onset of the various global work from home mandates. Therefore, many workflows and processes had to be adopted or rethought to incorporate remote access and automation of lab equipment. This, in turn, allowed scientists to continue their research without physically being in the lab. This led to organizations accelerating their implementation of cloud based applications and solutions that could subsequently be easily maintained remotely by IT staff and also allowing that remote access by the organization scientists that need to keep their research advancing and new and different directions. This focus on internal systems led by both business and IT to push the limits of technology, also helped organizations to see the gaps in their data flows and processes.

Furthermore, the data infrastructure that existed within these organizations have historically been built to enable specific answers to specific scientists’ questions. In other words, data has been stored in a way that only allowed for human points of access or interface. The emergence of machine learning or ML requires not human points of access, but machine points of access. And in order to achieve such machine access, it’s required data stakeholders to carefully consider the entire data infrastructure.

Subsequently, these organizations found out very quickly that they lacked the skillsets to take on any AI or ML initiative due to the challenges related to data standardization and harmonization. Along with the shortage of data scientists to help them better organize and manipulate their data to support such initiatives. So clearly these things have all accelerated various aspects of digitalization, automation, and data standardization.

06:06  Sarah Srokosz

Yeah. So to kind of build on that a little more and I think you got there in your previous answer, but part of our core customer base, like you mentioned, are the scientists or even more specifically, analytical chemists. And they definitely make up a large part of the end users of our software. So what do you see these analytical chemists specifically being interested in, in the year ahead? And is this in line with what you would have predicted three years ago, back in 2019?

06:36  Mark Meyers

Well, yes. Our core customer base is indeed analytical chemists, and the team members they support. Their work provides the fundamental basis for all types of decision making within their organizations. So these analytical chemists continue to look for ways to run more experiments in less time, while producing accurate and timely results. I believe they will continue to be interested in the automation of all types of workflows and data flows by continuing to seek out better ways to marshal data from instruments to result conclusion, with limited scientific intervention. Driving the analytical community to only review results that are outliers or are an exception to specific expectations.

And what I mean by that is the analytical workflows ACD/Labs specializes in, will be automated to the point that the scientists can rely on all calculations and computations to be accurate and correct with the algorithms in place, and they only need to perform their review with specific data or experimental results by exception.

This means all routine data processing can be automated and validated once without any manual intervention, allowing for a significant savings of analytical scientist time. In addition, in order for AI and ML initiatives to take hold, especially in the analytical space, data standardization needs to occur. ACD/Labs has been encouraging organizations to standardize on their analytical data formats for years, and organizations are finally starting to take notice and implement such capability.

ACD/Labs, in my opinion, is one of the few, if not the only organization uniquely positioned to help customers achieve this vision. As we support a variety of instrument types, vendors, or vendor agnostic, over 155 different instruments and file formats of analytical data formats today. There are also many organizations that are attempting to drive standards that ACD/Labs also support. These include Allotrope’s ADF format, their ASM format, which stands for Allotrope Simple format, their ontologies as well as support for other competing analytical data standard formats like AniML and JSON.

08:42  Jesse Harris

So going back to some of your previous comments, you mentioned a few times now already like algorithms and things like machine learning and there’s a lot of other buzzy topics right now. Automation, I think you mentioned already high throughput experimentation is another one, and cloud computing. Are these, is this excitement in these sorts of topics being translated into implementation? And if so, what does that look like?

09:05  Mark Meyers

Yeah, excitement, buzzy. I’m not sure I would call them all of those, but they are real and they deserve attention. And in many cases, they’re not new. There is just a renewed focus on them, whether it’s excitement around certain ones or not. They are indeed being translated into implementation today. For example, the use of automation and robotics and drug discovery has been around since the mid-nineties, as has the concept of neural networks and to some degree, AniML.

However, like any of these technologies, they have to be molded to fit specific use cases and solutions to be valuable. Initially, people try to apply these concepts to a lot of different areas, and they typically fail because it’s either the wrong use case for the technology, or the infrastructure, or the application. I personally have seen a lot of that over the years.

Therefore, for the concept to be viable, it also requires complementary technology components to go along with that to make them viable and adopted into the mainstream. ACD/Labs has been utilizing AI-type algorithms, and to be more specific, neural networks, since the mid 2000s. So to ACD/Labs, this is not necessarily new technology as it’s been embedded in various aspects of our prediction software and capabilities for many years. So not a new concept in AI, but people are again paying attention because the underlying technology is advanced to the point that utilization of AI in different areas is actually possible and valued.

In addition, the ability to automate workflows and data flows in the lab combined with the ability to harmonize and standardize on all types of data, are allowing the convergence of multiple technologies to create solutions in support of drug discovery and development, but in other aspects of the health care industry, as well as clinical trials, real world data analysis and the general health care industry.

Now to circle back to two other areas of your question, high throughput experimentation and cloud computing. For high throughput experimentation, many organizations, but certainly not all, have embraced the concept of chemists running more diversified experiments in a shorter time. The only way this can be achieved is to parallelize efforts, and instead of running one experiment at a time, perhaps running three at a time or 24 at a time, 96, 386, etc., and varying any number of experimental parameters, allowing simultaneous optimization of both starting materials, synthesis, protocols, catalysts, etc., all to improve the speed of real identification with potentially harmful impurities.

We see a lot of this in development, particularly in support of QBD, just one example. You can do this by exploring the chemical and process parameter space around an API synthesis or formulation, thus forcing an increase in the number of experimental runs, but also decreasing the time to result. And time is critical in drug discovery and development. And you can only accommodate such a large experimental scale by utilizing robotics and software designed specifically to manage the entire end-to-end workflow.

Fortunately, ACD/Labs spotted this trend several years ago and decided to implement a product called Katalyst D2D (Design to Decide) for many of its existing customers. And finally, cloud computing is here to stay. But I would caution organizations that not all applications are meant for the cloud.

12:29  Sarah Srokosz

Yes, certainly, I would agree with that last statement for sure. But we talk a lot about the pharmaceutical industry here; as you were mentioning, drug discovery and APIs. But we also support customers both in academia and the more broad chemical industry. Have you seen any unique trends or stories in any of those areas?

12:51  Mark Meyers

And you’re right, yes. ACD/Labs supports over 3000 plus organizations worldwide and tens of thousands of users in all different types of industries from academic, government, forensic, specialty chemical, consumer products, and the list goes on. Anywhere someone needs to perform any type of chemical analytical tests, ACD/Labs is there to support it with either its expert desktop tools or its enterprise tools.

These enterprise tools allow organizations to store, manage, process, query, and report on all different analytical data types. These industries, like the pharmaceutical industry, are generally following the same trends, each attempting to focus on automation, cloud computing, analytical data standardization, digitalization, and in some cases, high throughput experimentation, all in their own and unique way.

And currently, we’re working with several specialty chemical organizations in providing automated analytical data processing, capture storage and retrieval system, and support of their global analytical efforts and departments; not unlike what we do for our large pharmaceutical companies. So this ability to actually database analytical information and all the processing metadata is critically important to every industry that we serve.

14:07  Jesse Harris

Okay. So a lot of the same trends, it sounds like for sure. But generally for ACD/Labs, 2022 has been a very busy year. I am sure that you can probably agree with that, that you’ve had a lot on your plate. But I wanted to ask you if there any accomplishments or highlights from the year that you could share with us from the company or any important lessons that you took away from 2022?

14:31  Mark Meyers

Happy to. 2022? Yeah, it’s been a very busy year for ACD/Labs, variety of different reasons. ACD/Labs continues to have success in advancing its products and solutions and creating value for its customers in the industries that we serve, which is the most important thing we can do to maintain our longevity. Most notable are the continued acceptance of our enterprise solutions, including Katalyst D2D, which I mentioned earlier, Luminata, our CMC Decision Support Tool, and our enterprise analytical data management systems; as well as a continual delivery of our robust expert analytical desktop tools and predictive software capabilities.

Also, ACD/Labs recently embarked on a journey to migrate its existing expert desktop tools to the web, a journey that started last year. And at the end of 2022, we have made significant advances in fulfilling our vision by adding the capability to process LC/MS data on the web, with more capabilities to come in 2023. And to continue, one of the more interesting highlights for us, is the most recent announcement of our continued and longstanding partnership with CAS Chemical Abstracts Services.

CAC has, over the last 20 years invested in ACD/Labs NMR prediction algorithms and our other Physicochemical property offerings. More recently, they made an additional significant investment to access additional property predictions, both in ADME/Tox along with revised NMR predictors to enhance their own product offerings. CAS is a respected source in this space and continues to use and see ACD/Labs as a premiere supplier for these algorithms and toolsets.

Finally, the one important lesson from 2022, like any other year, is that providing important, valuable tools for our customers, whereby they continually invest in our offerings, is the most rewarding experience an organization like ACD/Labs can have. And we hope to continue to do so in the years to come.

16:31  Sarah Srokosz

Yeah, that sounds like a very busy year for sure. And so as someone who is right in the thick of all of this, I’m really interested to hear what are the trends or topics that you’ll be following to start out 2023?

16:45  Mark Meyers

Well, I think I mentioned them. They continue to be persistent in the industry. And so automation, machine learning, and analytical data standardization and data formats will certainly be at the top of the list for sure. And so we will continue to make significant advances in those areas. And in our web-based expert toolset.

17:07  Jesse Harris

Great. Thank you so much for giving us a little information and a very wide view of the field. So that’s been lovely to have you and get all of that context. It was very interesting.

17:20  Mark Meyers

Well, thank you, Sarah and Jesse. I appreciate the opportunity to do so.

17:25  Sarah Srokosz

I’m so glad we got to hear Mark’s take on the various trends in the world of chemistry and chemical data. It sounds like there’s a lot to be excited about.

17:33  Jesse Harris

Absolute. And that is good news for us, as we’ll have plenty to talk about in 2023.

17:39  Sarah Srokosz

Indeed. If you’d like to learn more about our partnership with CAS that Mark mentioned, there’s a link to the press release in the show notes.

17:47  Jesse Harris

Thanks for listening to the analytical wavelength. Best wishes over the holidays and we look forward to bringing you more conversations in the New Year.

Analytical wavelength is brought to you by ACD/Labs. We create software to help scientists make the most of their analytical data by predicting molecular properties and by organizing and analyzing their experimental results. To learn more, please visit us at www.acdlabs.com