In 2024, our role as a trusted partner to our customers has been front and center. As we shift into the new year, our focus remains on delivering tangible scientific product innovation and value across all our product brands. With the addition of Dotmatics Luma, we’re not only breaking down data silos, but we’re also enabling seamless interoperability and powering comprehensive, scientifically precise, multimodal workflows. These are critical steps for our customers to fully harness the transformative potential of AI-driven discovery.
We know that our customers are doing more with less resources these days. This theme has continued over the last few years of pandemic recovery efforts. Efficiency remains key. AI has the promise to play an increasingly important role in freeing up time and resources, yet it’s not the full story. Because although companies are searching for new tools to help them, they’re also getting creative internally—finding ways to reallocate existing resources, often replacing technologies and cutting costs to open up more budgets.
Overall, there are three meta trends we’re hearing most frequently as being mission critical to our customers.
Trend #1: To make R&D more efficient, companies need a Lab-in-a-Loop
We know that for AI to create meaningful impact in life sciences, the drug discovery process must evolve to become a Lab-in-a-Loop, where R&D data and clinical data across applications, databases, and lab equipment are ingested, centralized, and then used to create models of prediction for the next set of experiments. I think most scientists agree. The goal is an iterative cycle. AI models, trained on R&D and clinical data, can predict and refine experiments, creating a more efficient interplay between the wet lab where experiments are physically performed and a dry lab where AI predicts which experiments will be most successful, or even predicts the form therapeutic molecules should take. The result is a faster and more efficient drug discovery process.
Today, there are many different terms for platforms that strive to fill this need, and it’s early enough that there is no consensus on terminology. Some experts call it a Unified Lab Informatics Platform (ULIP), or a multimodal scientific intelligence platform. Regardless of nomenclature, such a platform is essential for modern labs because it brings together scientific precision in molecular representation and lineage, data management and processing, specialized scientific software, and adaptive workflows into a single, streamlined system. It integrates data from diverse sources, eliminating silos and enabling seamless collaboration among teams. It boosts productivity and ensures faster, more accurate decision-making by automating routine tasks and providing real-time ingestion, contextualization, and access to harmonized data.
Since we launched Dotmatics Luma in 2023, followed by Lab Connect, the Luma data and instrument ingestion engine, earlier this year, we’ve seen tremendous interest and activity using the platform. We already have more than 2,100 connected instruments that have parsed over 47 billion data records and 300 terabytes of data. That’s massive. And as more and more labs embrace the potential of AI-driven discovery, Luma will serve as a future-ready foundation, supporting their innovation and scalability.
One major multinational pharmaceutical company set a goal to deploy Luma to 1000 instruments within a year, and within the first six months had already connected 2000+ instruments—4X faster than expected. In fact, within 10 minutes of seeing Luma at work, team leaders said they realized it was categorically different than any other solution on the market. That’s because of its seamless linkage of the output processing with the extraordinarily flexible, yet well-governed data model, its advanced dataflow and data analysis capabilities, and its quick deployability.
Similarly, the oncology R&D group at a top US pharmaceutical company recently deployed our Luma Flow Cytometry Workflow. It took Dotmatics less than 10 hours to remotely deploy an environment to parse the team’s data using Luma Lab Connect. Rollout was also fast, with five instruments and 20 users connected and onboarded in one week. Each scientist is now saving multiple hours every week thanks to better connection of instrument files to flow cytometry software, plus the countless development hours saved by putting the brakes on a costly home-grown solution.
Trend #2: Multimodal discovery is becoming a reality
Companies are increasingly adopting the approach of addressing medical needs using whichever mode of action proves effective. This means they need a truly multimodal informatics system now more than ever. They need tools that can handle conjugate therapies, that can tackle target-focused research regardless of the applied therapeutic modality, and that provide a more robust and economical solution across the therapeutic modalities due to not needing to stitch together multiple disparate solutions.
Because multimodal research is inherently more varied, the platforms that can support it must be more flexible and extensible. Developing a platform with flexibility and extensibility in mind lets us and our customers adapt our software to new, as of yet undiscovered modes of research and molecule types of tomorrow. The platform reflects and enables how scientific discovery actually takes place.
Because of this, our newly introduced Geneious Luma can support any sequence-based modality—monoclonal antibodies, multispecific antibodies, antibody drug conjugates, CAR-T cell discovery, siRNA, CRISPR therapeutics, and vaccine discovery. Geneious Luma enables researchers to use the advanced bioinformatics, molecular biology, and antibody discovery capabilities of Geneious Prime and Geneious Biologics to design, qualify, annotate, and filter sequence, and assay data. Seamlessly working together with Luma, researchers can execute their cloning, expression, and purification tasks using Luma’s Adaptive Workflow capabilities.
Multimodal biologics research is divided into two bins: conjugates and target-based discovery. Multimodal research has expanded the horizons of protein engineering. This approach integrates both a confluence of datasets and methodologies, and requires a platform that is capable of supporting a variety of therapeutic modalities. In the context of antibody engineering, multimodal research includes both conjugate development and target-based discovery.
Conjugates involve combining different research modalities to create novel molecules, while target-based discovery focuses on identifying the best modality for a specific target within a unified platform. Conjugates enable mixing different modalities of research to arrive at a combination molecule. Target-based discovery, which focuses on identifying the most suitable modality for a specific target, requires an integrated platform to streamline decision-making. Without a unified system, researchers must bounce around different data sets and software tools. Both conjugate development and target-based discovery are heavily reliant on DNA/RNA/protein sequence analyses throughout the R&D process and final protein production, which is where Geneious Luma is a leader in the biotech space.
Multimodal research platforms reduce the need to cobble together disparate systems. When paired with a low-code environment that creates greater business agility because business and IT can collaborate iteratively on outcomes. With Luma, both scientific R&D and business data science can take place using one platform. And because scientific data requires context and lineage information to rationalize, capturing it in situ within a unified platform, means spending less time on expensive downstream tasks such as data collating and cleansing…where it’s often too late to identify the appropriate context.
Importantly, multimodal research might open up the ability to train models on outcomes that span modalities; it’s foreseeable that you could predict the efficacy of a particular type of molecule against a target based on historical data across all targets and modalities under various conditions. This would allow researchers to gain deeper insights and make cross-disciplinary connections that were previously impossible.
What’s exciting with Dotmatics Luma is that in addition to being a platform for multimodal discovery, it addresses the needs of biologics-based research—the fastest growing research area—far better than anything available to the industry today.
Trend # 3: AI at the tipping point: driving tangible outcomes
Two years ago the AI hype cycle created such a frenzy, but there wasn’t a lot of clarity on how and when the technology might impact R&D broadly. Today, the intersection of AI and life sciences is no longer a futuristic concept; it's happening now. Our customers are at various stages along the path of transformation, and are looking for the right tools to help. There is a strong incentive to introduce new resources that put the power in scientists' hands to decrease the burden on data science and IT, and significantly expand the value of AI/ML.
Likewise, multispecific antibodies provide a big opportunity to develop innovative treatments that improve patient care for complex diseases like cancer and autoimmune disorders. For that to happen efficiently, researchers will need tools enabling the design and generation of multispecific antibody formats and the execution of predictive ML models to score potential multispecific antibodies, so that they can proceed with only those that are most promising. This is why we recently introduced the Luma Antibody & Protein Engineering solution, a comprehensive end-to-end solution for streamlining the antibody R&D process, with an emphasis on monoclonal and multispecific antibodies.
The pinnacle of the AI journey is "Composite AI," where scientists leverage all their data across disciplines, including cascading layers and multiple types of AI, to drive simulations, predictions, and novel recommendations. For example, scientists can leverage auto-gated flow cytometry results produced in OMIQ that are then used alongside other assay data to train down-stream models for molecular liability calculations. It’s this coming together of multiple types of AI-processed or predicted data where the true potential of AI for drug discovery is realized, and where Dotmatics Luma plays its most essential role.
This is also where big data is critical. The rule for AI/ML is that you need a LOT of data before the modeling can be useful. The increasing availability of data from many sources is driving biotech research, helping to find and interrogate old and new targets, to refine treatment approaches, all toward improving patient outcomes, especially as patient data are related to experimental data. That means seeking out smart integrated data strategies that enhance an organization's existing software spend. In the coming year, customers must focus on improving their data strategies to boost their AI adoption and effectiveness.
Finally on the regulatory front, the introduction of AI model governance will become a major customer priority in 2025, and beyond. We need to ensure models are accurate, ethical, and compliant with regulatory standards, to reduce risks of bias and errors in drug development. This is crucial for maintaining data integrity, patient safety, and public trust in AI-driven advancements in drug discovery. There is also increased importance on traceability of data, especially in the design of new proteins. This includes the need for lineage tracking back to in silico designs.
We’ve got a ton of work to do. But as we head into 2025, we’re excited to partner with our customers to tackle these trends head-on—breaking down silos, enabling smarter workflows, and driving discovery faster than ever.
The future of science is bright, and we’re just getting started.