Co-authored by Plamen Petrov, Discngine’s Senior Software Architect

Data management, comprehensive analysis, and efficient workflows are just some of the fundamental features of informatic tools for plate-based assay campaigns. Indeed, having a robust application system to manage data from screening experiments enables lab teams to improve lab operations' efficiency and focus on their next best decision.
Therefore, drug discovery research laboratories largely invest in the license and implementation of such applications and use them to simplify their daily work.

Based on the Cell-based Assays Market report by Bloomberg, the global cell-based assay market size is expected to grow up to USD 33,8 billion by 2030, with a compound annual growth rate (CAGR) of 8,75% (from 2022 to 2030).

Click to zoom in: Based on the Bloomberg report, the global cell-based assays market size will almost double in around 10 years, reaching USD 33,8 billion by 2030, with a compound annual growth rate (CAGR) of 8,75% (from 2022 to 2030). Link here

Although these projections are subject to change due to various factors (market dynamics, technological advancements, regulatory requirements, etc.); there is a trend that reflects the scientific community's collective recognition of the need for efficient data handling and analysis to accelerate the pace of discovery and development.

But what makes a good assay data management system? How do you make sure it scales with your company?

In this blog post, we highlight five criteria for a healthy plate-based assay data management system, going beyond fundamental requirements (e.g., fast, efficient, easy to use, etc.).

1. Science-native design

An important criterion to consider for a healthy assay data management system is its software architecture. This means that the assay software should be specifically designed with “scientists in mind”, rather than having generic data management features repurposed for research. By following the way scientists think, these systems account for multiple data formats, cover various biochemical, biophysical, cellular, and other assay types, analyze complex data sets across all assays, and more. Although not necessarily easy to use, those systems should be intuitive enough for scientists in the lab that are usually not IT experts.

For example, software should have a simple interface with minimal and clear indications for proceeding to the next step. It should avoid squeezing too many features onto a single page, making it challenging to distinguish between basic and advanced options.

It is worth noting that identifying such systems is challenging since many different tools vary in terms of features and capabilities. One effective approach to recognizing them is by considering their origin – assay systems designed by individuals with dual backgrounds in both science and IT. The developers’ deep understanding of the underlying scientific principles empowers them to create an interface that is both intuitive and tailored to the needs of science-based users. By bridging the gap between scientific knowledge and technological implementation, they ensure that the resulting interface is not only intuitive but also optimally supports scientific endeavors.

All in all, assay software that is science-native provides research-relevant results faster and enables confidence in the accuracy of research results, leading to better decision-making in drug discovery.

2. The Big Picture

Software with features adapted to the scientific mind should also account for evolving scientific needs. One side of this “big picture” criterion includes the system’s ability to adapt to changing research demands over time, emphasizing its flexibility in accommodating evolving methodologies and requirements. It is important to ensure the system you implement initially can be upgraded later with features that scale, such as new modules, custom codes, plugins, etc. Otherwise, with every new requirement, scientists would have to switch to a different system or develop entirely new software, which is inefficient and time-consuming.

For example, more and more lab teams are adding artificial intelligence (AI) features to run machine learning models on assay data to analyze trends, and make predictions. Or maybe they wish to accommodate an in-vivo module or any other assay they can develop, regardless of its complexity. If the system is flexible, emerging features can be added easily, extending its usage and user confidence.

Another side of this “big picture” is the scientific needs related to data visualization. Lab experts look at visual data instead of numbers. Therefore, to get a view of all the data at hand (historical and current), the system needs to have powerful visualization that enables the seamless analysis of HTS screening campaigns, or cross-experiments and multi-assay campaigns. The latter allows users to identify patterns and trends over time that might be otherwise difficult to see. Not just now but also in the future.

3. Right people, not just the right license

Having access to knowledgeable support is paramount, surpassing the significance of even a valid software license. As early as in the evaluation phase of the new assay data management system for your lab, the criteria to have in mind are the reliability and responsiveness of the vendor company. From initial steps like software implementation and installation right throughout its day-to-day usage, your team will naturally encounter questions and technical problems:

“How can I read files from my new instrument? How can I change the curve-fitting model?”

Thus, if your system doesn’t have a dedicated scientific and technical team to understand research needs and provide timely aid with issues, productivity in the lab will drop significantly.

The right people alongside the software licence come with:

Scientific and technical background that allow them to understand scientific needs of the users
Hyper-vigilance for questions and scientific/technical issues
User documentation and training in many forms

It is also important to keep in mind that data management software are often complex and not intuitive to use. Especially at the beginning, users need to go through the learning experience to familiarize themselves with technology and features. That’s why it is important to work with a vendor that provides proper training and resources as a core part of its support. Since one size does not fit all, the support should come in different forms - written documentation, cheat sheets, video tutorials, descriptive manuals, and anything else that helps users get up to speed quickly.

By prioritizing strong customer support, a healthy assay data management system can help ensure that researchers can focus on their work, without getting overwhelmed by technical details.

4. Data warehouse and beyond

Another important capability of a healthy assay data management system is that it should have a data warehouse. Many scientific discoveries and decisions are based on assay data, frequently scattered across organizations and external collaborators like CROs. If each individual handles their data separately, keeping track of data provenance can be challenging, leading to an increased risk of errors and duplication of work. Showing that, it is worth having a system that stores, structures, and organizes all assay data, including the historical data and annotations, in one centralized repository. With a single source of truth for assay data, data inconsistencies can be avoided.

In addition to the warehouse, a data query system should also be set up to help process and analyze data and get the expected benefits of storing it. It takes a lot of effort and time to acquire assay data; therefore, an important characteristic of the data management software is to enable the entire data journey – from data capture to data retrieval. Moreover, such a system helps achieve FAIR data, which is the pursuit of many biopharmaceutical R&D. Accessible, manageable, and reusable FAIR data facilitates data sharing and effective communication regarding data management and interpretation.

All in all, a searchable data warehouse with a query system can give researchers easy access to the data they need, allowing them to make better data-driven decisions.

5. The power of synergy

Finally, it is important to highlight that it will not always be possible to get a system with all the criteria we listed above or with all the desired features. When this happens, the solution is not to settle for an inadequate system. Instead, experts should consider the power of synergy by combining the strengths of multiple software and integrating them into one robust assay data management system.

For instance, one lab software might have excellent scientifically designed workflows with flexible built-on capabilities but lacks a superior data warehouse with a query system. With powerful APIs and integrative abilities, the systems connect and fulfill each other’s gaps. With the synergies between big systems, there are no tedious, copy-paste manual activities that are highly error-prone and inefficient.

Therefore, integration is an essential criterion of a good assay data management system, even if some other features are missing. Since one size doesn’t fit all, combining multiple software lab experts get custom-tailored solutions that meet the unique needs of their research lab.

Summary

With the rapid growth of screening-based campaigns and data-driven drug discovery, there is an increasing demand for efficient assay data management tools. We have identified five key criteria for a healthy software:

Science-native workflows
Holistic view or big picture capability for future experiments and complex visualizations
Adequate scientific and technical support from the vendor company
Data warehousing and querying capabilities
Seamless integration

What are your thoughts on these criteria? Do you agree?

Feel free to share your opinions in the comment section below.

Five essential criteria for a healthy assay data management system