Bridging Scientific Intelligence Across The Product and Process Lifecycle

Dotmatics Briefing Note

Axendia was briefed by Steve Tharp, President, and Kalim Saliba, Chief Product Officer, at Dotmatics, a Siemens company. Dotmatics provides R&D software and data management solutions to researchers across life sciences, chemicals, and materials. The AI-native Luma platform is central to Dotmatics and Siemens bridging scientific discovery and scale-up – across the product and process lifecycle – and enabling a digital thread across ecosystems that have historically been fragmented and siloed.

Fragmented Data Across a Long and Costly Pipeline

The life sciences industry faces a well-documented challenge: it still takes 10 to 15 years and billions of dollars to bring a drug to market, with research comprising roughly 30% of that duration. A persistent contributor to this timeline is data fragmentation, i.e. scientists working in siloed systems, CROs returning incomplete data sets, and critical process context being lost in handoffs among teams and tools. Another hindrance is that much of the data also needs to be structured and contextualized for teams to apply AI, ML, and advanced analytics.

Tharp framed the opportunity plainly: “As we know, 95% of candidates don’t actually succeed past R&D, and 99% of experiments fail. So, if you can have all of those processes of what was done, why it was done, and what data did it create, proximate to each other – you now have the training ground that AI actually needs to be agentic, not just generative.”

Saliba reinforced this, describing customers that are “lacking an overarching platform that can both support the creativity that’s required by early-stage research, and the structure that’s required to piece it together in a way where it’s actually useful – not just to their departmental colleagues, but also downstream in tech transfer and manufacturing.”

The implication is significant. Without a structured, traceable record of scientific work – including what failed and why – AI models lack the context required to act meaningfully as co-scientists. Per Tharp, Dotmatics’ position is that solving this data problem is a prerequisite to any meaningful AI deployment in research.

Scientific Applications Meet Enterprise Infrastructure

Dotmatics operates two primary lines of business. On the scientific applications side, the company supports a broad portfolio of tools familiar to bench scientists, including GraphPad Prism, SnapGene, and Protein Metrics, which span statistical analysis, molecular biology, and proteomics. These tools serve as sources of high-value scientific IP and are used broadly across academia and industry.

Saliba acknowledged that some commonly used tools have also contributed to the very problem Dotmatics is attempting to address: “We are well-positioned – and frankly, somewhat accountable – for helping to solve it. We have acquired a lot of those applications and are therefore very much responsible for reducing much of the silo proliferation.”

On the enterprise side, Dotmatics offers its Electronic Lab Notebook (ELN), the Luma platform, and Virscidian, a high-throughput purification and small molecule automation solution. The portfolio supports small molecule, large molecule, and cell and gene therapy workflows, and is designed to serve multimodal research programs focused on disease rather than any single platform modality.

Tharp noted that Dotmatics is actively building out capabilities to deeply support increasingly complex modalities: “We have the ability to support ADCs. The whole platform is multimodal, and we’re going to keep bringing our scientific expertise into Luma – not just high-level structure search, but built-in chemical intelligence, as well as everything around protein folding, cloning, and expression.”

LUMA: A Plaform Built for AI-Native Science

The centerpiece of Dotmatics’ enterprise offering is Luma, a platform designed to bring together scientific data, process context, and AI capabilities in a single, scientifically aware environment. Tharp elaborates that “Luma is not a generic data lake or laboratory informatics overlay. It’s built around the concept of deep data registration, where every experiment, material, process step, and result is captured with enough scientific precision to support downstream AI analysis and decision-making.”

Saliba described Luma’s role as a low-code app-building layer that orchestrates workflows and dataflows across scientific functions and departments – replacing the patchwork of siloed ELN systems, LIMS systems, instrument integration pipelines, and separate AI/ML environments that organizations have historically cobbled together. The goal, as he put it, is to bring “all facets together at the point of use instead of after the fact” – in contrast to the retroactive approach of collecting data from disparate systems and then attempting to derive meaning from it.

Luma intends to enable scientists and IT teams to define apps and adaptive workflows without requiring custom development for each use case. Saliba noted that customers are increasingly “vibe coding” – building applications using AI-assisted tools. While the energy around this trend is real, so are the risks: “The biggest challenge is governance. Each separate app is going to consume and produce data, and if those datasets don’t conform to an ontological consistency within the rest of the organization, forget about connecting meaning between those applications. Luma provides the structure within which these innovations can happen safely and sustainably.”

A key architectural principle is persona-based access: scientists see what is relevant to their role and modality, whether that’s a protein engineer working in protein design or the head of biologics looking across programs for a status report. Tharp described the aspiration as giving every scientist “a co-researcher sitting next to them in Luma.” This assistant can help with repeated tasks, query historical data, and eventually operate with greater autonomy.

Saliba elaborated on how Luma’s AI capabilities differ from generic large language model (LLM) approaches. Rather than relying on retrieval-augmented generation (RAG), Luma Agent which is MCP-enabled and can be surfaced within tools like Claude – shows users the actual queries that are being run to answer their questions. He also highlighted the role of domain-specific AI: purpose-built machine learning models embedded within individual scientific applications, which deliver targeted analytical value more efficiently and cost-effectively than a frontier LLM could.

Luma’s Lab Connect capability extends this further, enabling instrument-agnostic data ingestion across thousands of instruments and systems. Tharp described one large pharma customer processing over five terabytes of scientifically registered data per day across over 3,000 connected instruments. This equates to upwards of 500,000 files per day. Billions of data points can be searched, traced, and used to defend decisions in regulatory audits years later.

Bridging Discovery and Manufacturing: The Siemens Dimension

A defining aspect of Dotmatics’ current trajectory is its position within Siemens. Tharp described the complementary nature of their respective offerings: Dotmatics has historically addressed the portion of the drug development timeline associated with research and discovery, while Siemens more broadly covers the remaining timeline associated with process design and manufacturing.

Bringing these solutions together creates the potential for bi-directional data flow between research and manufacturing and to streamline tech transfer, a challenge that Axendia continues to hear from the industry. Tharp summarizes that “If you have all the experiments, the materials, the processes, the steps, and the whole recipe on how you got to the molecule that is being advanced – and you can simulate and automate the scale-up processes – then you can significantly reduce the hurdles, the paperwork, and the data fragmentation that hinders scaling up.” Three Lighthouse customers are actively co-developing this capability with Dotmatics today.

In Brief

As a tool, AI has tremendous potential to improve efficiencies across the product and process lifecycle, in R&D, and during tech transfer. The emphasis on traceability, scientific context, and open data registration (capture and formal recording of scientific entities) distinguishes Luma Agent from platforms that simply layer AI on top of existing data stores without addressing the quality and structure of that data. As life sciences organizations continue to invest in autonomous lab and AI co-scientist initiatives, the availability of a platform that can scale from research to manufacturing has the potential to add significant value. Weaving Dotmatics capabilities within the Siemens portfolio can be a meaningful differentiator.

We will continue to provide updates on Dotmatics as they become available.

Related Content:

Stay Connected with Axendia!

Stay at the forefront of innovation and technology in life sciences, healthcare, and space-enabled research by connecting with Axendia.

Follow us on LinkedIn: Join our professional network and stay updated on the latest insights, trends, and case studies. Connect with Axendia on LinkedIn
Subscribe to Our Updates: Get exclusive insights and thought leadership articles delivered straight to your inbox. Sign up for our newsletter
Share Your Thoughts: We’d love to hear from you! Contact us to discuss your ideas, challenges, or opportunities. Email Axendia

Let’s drive innovation together!

Ready to unlock new opportunities and drive success? Schedule an Analyst Inquiry

The opinions and analysis expressed in this post reflect the judgment of Axendia at the time of publication and are subject to change without notice. Information contained in this post is current as of publication date. Information cited is not warranted by Axendia but has been obtained through a valid research methodology. This post is not intended to endorse any company or product and should not be attributed as such.