video

Breaking Down Data Silos: Accelerating Biologics with Scalable Platforms (Podcast)

With Veronica DeFelice, hosted by Sapio Partner Csols.

In the latest episode of Accelerating Biologics with Scalable Platforms Beyond the Data Silos, Veronica DeFelice explores how life sciences organizations can overcome one of their biggest hurdles, data fragmentation.

The Core Challenge

Biologics research generates massive, complex datasets spread across disconnected systems. This not only slows discovery but also limits insights. Veronica highlights how a science-aware informatics platform, like Sapio’s, can unify data, streamline workflows, and ensure compliance, unlocking faster, more informed decision-making.

Key Takeaways

• Digital Twins & AI: Creating virtual models of lab processes enables reproducibility and predictive analysis, while AI tools like Sapio ELaiN empower scientists to interact with data using natural language.
• Unified, Scalable Infrastructure: Moving beyond legacy systems to a configurable, cloud-ready platform breaks down silos and accelerates therapeutic development.
• Actionable Insights at Speed: By simplifying data access, labs can move from experiment to insight more efficiently, supporting both research and clinical workflows.

Why Listen?

If you work in biotech, clinical research, or lab operations, this podcast offers a practical roadmap to digital transformation. Veronica provides real-world strategies for unifying lab data, integrating AI, and building scalable systems that keep pace with innovation.

Watch the Full Interview:

Read The Full Interview

Lisa Richard
Welcome to Decoding the Digital Lab brought to you by CSols Incorporated. I’m your host, Lisa Richard. And today we’re happy to have Veronica DeFelice, a biologic subject matter expert with over a decade of experience, join us as the Director of Biologics at Sapio Sciences. Veronica and her colleagues are building solutions that help organizations seamlessly connect vast amounts of scientific data to help accelerate the delivery of treatments for patients. Today, we’re going to drill into the challenges of precision and tracking of biological data and overcoming these through the deployment of scalable platforms. Veronica, welcome. It’s a pleasure to have you.

Veronica DeFelice
Thank you so much for having me, Lisa. I appreciate it.

Lisa Richard
Of course, we’re glad for it. Why don’t we get things started off with your view of the biological complexities with data integrity that scientists are facing?

Veronica DeFelice
Yeah, I think the biological complexity with data integrity involves a lot of systematic tracking and molecular modifications and variations while connecting laboratory experimental outcomes specifically with the computational predictions. And that encompasses a lot of the impact of genomic and proteomic and also gene expression data with the exact research outcomes. So to be able to have that—the complexity of that data—be transferred forward and illuminated on, we can really see that in different tracking mechanisms to ensure data integrity.

Lisa Richard
When we think about the challenges that they face, is it in the execution of the experimentation? Is it in the uniqueness of the data? Is it in the sort of collation of it—where is kind of the crux of the challenge that they face?

Veronica DeFelice
I think, from my observation—specifically being a scientist and having that as my background—I think that the most difficulty that we see in a lot of different client conversations and also in my prior experience really roots in data silos. Not really the experimental execution, but being able to connect the data together and build experimental conclusions that have a sense of reliability and traceability, and to be able to draw those insights that are very foundational to building a therapeutic.

Lisa Richard
So in looking at the challenges that they face—and I think you mentioned a little bit about data silos—and so when we look at biologics in particular and in sort of where we head with platforms, talk about those challenges and how they’re kind of overcome from a collaboration perspective, or elaborate further on what those challenges are from that perspective.

Veronica DeFelice
I think specifically in a unified laboratory platform where collaboration really can eliminate a lot of the issues with not being able to see the different results per process. And with a platform that’s centralized and collaborative, you’re able to look at the different types of results in reference to the different companies—even outside of your company—that might be needing the samples, or even internally, whether it’s a biologics group or a small molecule group or a group working specifically in analytical characterization or more process development. In a digital sense, you can see the entire research platform and how the integrated laboratory informatics can be transferred from each point. So it becomes comprehensive, but also can get detailed once you get into each of the results, and even how the experiment was conducted can be illuminated in the digital platform.

Lisa Richard
Yeah, I’m thinking of some experiences that I’ve had where that’s one of the biggest challenges, right? So it’s the respect for the scientific needs of each science—biologics, chemistry—each of them, but also the respect for the collaboration that’s needed, right? How do we share that information? How do we access it so that it can really drive decision making?

So in that vein, I think in the beginning we were speaking about the precision and accuracy of the data, right? Can you give a little bit of conversation around how that collected data analysis can actually drive more precision in the discovery process or in the decision making around it?

Veronica DeFelice
Yeah, I think that for example on biological data, there’s so many variations that occur. And to be able to ensure the precision and the accuracy throughout complex experimental workflows, you need to be able to go back and see the different modifications as well as understand the insights that you gather from each process output. So in a digital sense, we can have an input—let’s say about a biological structure—but an output would be about that function. And in the precision nature of it, you can really be able to weed out things that are inconsistent or flagged if something is not appropriate. You can analyze just for yourself, and that really sets in the scientists. And that also shines light on the scientists being in control, because the scientist can really be able to look at, let’s say, a sequence, and even though the digital platform is essentially analyzing it—for example, in multiple sequence alignment in a molecular biology tool—you can see the different sequence alignments and you see all the different annotations. But then I can go in as a scientist and I can make the decisions. If I say, “OK, this nucleotide, I really think has the point mutation,” then I can move onward. So it gives you a level of clarity, but not to the point where you override the decision making of the scientist. So I think that is a precision point. But the analysis and the different types of tools that are used to gather certain insights are really helpful to the precision nature of building a therapeutic.

Lisa Richard
I think that’s amazing. One of the things that resonated with what you said is a key challenge that I hear often with scientists is there’s so much data, and there’s so many permutations of that data as they move through the optimization process. Being able to decipher and quickly target which pieces of information will feed into the next decision, but still having that traceability—and I think that’s a huge value that you’ve kind of shined the light on, per se, of these platforms. And so I really appreciate that.

When we look at kind of a real case, is there any real case that you’ve worked on with the deployment of these platforms and the scientific value of outcomes? Take us through that so we can sort of see that in the real world.

Veronica DeFelice
Yeah, and I think that segueing off of the data infrastructure we just spoke about and really leveraging digital twins in a scalable way—for example, monoclonal antibody creation processes—when you optimize clone selection and scale-up efficiency through informatics platforms and then adapt to that research demand. For example, if there needs to be another library preparation step or there needs to be better expression data that comes, you can redo or have different assays be implemented. So it’s very, very adaptable. However, that’s a really good example of something that we could model in the platform. And you can build the monoclonal antibody from start to finish, and you can really illuminate each process that goes into it, and each experiment. But also, if there’s any different equipment that needs to be integrated, or their equipment software that needs to be integrated within that step, that can be very, very important and shine lights on the digital environment—where you can just pinpoint each part that you’re at and also work with your colleagues. I think that’s a big point as well.

Lisa Richard
So I want to zero in on one aspect. When we think of the art of science and the art of drug discovery, and then we want to apply digital systems, there’s this hesitancy and fear that “I’m going to be locked into a structured approach.” Can you speak a little bit from the scientist’s view of how flexible or customizable these platforms are to accommodate that constant pivoting that has to happen as you’re moving through the optimization process?

Veronica DeFelice
Yeah, I think that through being able to make decisions on each of the pieces, we really can, as the scientist, choose what module or what plugin you need. And we don’t change or force anyone to do everything in the workflow. It’s important that they have autonomy in what they feel is important for their workflow. For example, if an automated process doesn’t work for them and they really say, “I only need everything from library generation to purification,” or “I only want to look at high-throughput screening and continuous assay development,” or “I really want to look at just my plasmid or my vector modification construct,” that can be a choice and a selection for them. And we welcome that. I think that’s really important—to be able to choose, and then know that we also have these capabilities that can accommodate a whole end-to-end workflow. So it’s definitely very customizable to a process. For example, like monoclonal antibodies or even other things—like cell and gene therapy or looking at virtual cloning or just bioprocessing. There are a few modules that can stand alone, and so it’s important that we have a continuous process, but also be able to directly help something that we can accommodate in our ecosystem—in the platform.

Lisa Richard
Yes, that’s well said. I want to focus on the data piece that you mentioned, but I want to ask: do you find that these platforms—again, there’s the execution and the benefit to the operational use of these systems to aid in the process—and then there’s the benefit in the end, which I want to get into later, which is really how we use the data, right? Can we enable AI and use that to drive the decision making? But sticking to kind of the operational aspect and the need for the scientists to kind of evolve as they go, or do continuous monitoring, as you say—do you find that these platforms support both, or do you find that they really support kind of that end-value decision-making drive that we’re heading towards?

Veronica DeFelice
I think that digital platforms can really support continuous modification. And also, having an integrated platform that integrates very seamlessly into a scientist’s workflow makes them feel at ease, while also understanding that there is a lot of decision making—like you really shined light on—that the scientist has to make. Decisions on what can happen next, or what the next process should be, or if the data makes sense in a certain way or needs to be regenerated. I think something that a digital platform can really help with is it can help feel like the operations behind the scenes aren’t at the forefront of the scientist. So the scientist can really just focus on the science.

Lisa Richard
OK. I can see a lot of value. I’m sort of thinking in my head of lots of folks that I’ve spoken to that I would want to call and connect them to. “Oh, that’s so nice, you should do this.”

So we talked about the precision and accuracy and how it helps with that. We also talked about the collaborative need—and it’s not just biologics, it’s not just chemistry. There’s sort of a drive to putting it all together, and one of the key reasons for that is when you have all that data in a standard platform or a standardized format, you can enable the next value of technology, right? Everyone’s talking about AI and how we can use that to speed up outcomes or enable better decision making.

Can you shed a little more light on that and how you pull all of this conversation together to show how that can drive and benefit the use of AI on the data?

Veronica DeFelice
Yeah, I think that one example I could think of that really drives prediction of molecular behavior and also drug efficacy is for complex therapeutics like antibody-drug conjugates—which really have that biologic coupled with the small molecule—while enabling de novo protein design and different digital concepts that can help therapeutic candidates. So from a digital sense to a lab environment, I really think that AI amplifies scientific discovery by handling computational complexity and pattern recognition, like you said, while keeping researchers in full control of things like experimental design, hypothesis generation, and critical decision-making points throughout the therapeutic development process. So I think it really assists scientists, while the scientist is in control. [Learn more about Sapio’s scientific AI assistant ELaiN].

Lisa Richard
I think that’s a great way to put it. I was reading an article about the notion of AI enabling better decision making versus AI allowing you to make better decisions. At the end of the day, what you’re trying to do is increase the visibility and increase the understanding so that better decisions can be made—but it’s not making the decision for you. It’s helping you drive that. And so I think that was an excellent point.

We’re kind of running out of time, so I want to leave you with the opportunity to share what you want the listeners to leave with. What’s the key point you want them to take home about this notion of using scalable platforms to enable better biologics?

Veronica DeFelice
Yeah, I think that being able to have a scalable platform and do things with biologics, and be able to do each different process within biologics—but also be in control of the digital environments in reference to the wet lab and the science—I think holds a very special point. As we advance in different scientific domains—for example, in small molecule chemistry or in different other biologic environments like analytical characterization across development—the way that we can see things very clearly, and the lineage and the hierarchy of all of our different processes and referencing them back, I think is really important in developing a therapeutic.

Lisa Richard
Excellent. Thanks for joining us for this episode of Decoding the Digital Lab with CSols Incorporated. Join us next time, and hit the follow button if you’re enjoying this content. Thanks so much for joining.