Signals Xynthetica

Signals Xynthetica™ is an AI-augmented drug and molecule discovery platform from Revvity Signals that unifies in silico design with experimental science in a single, integrated environment. By bringing predictive models, experimental data, and laboratory workflows together, Xynthetica enables R&D organizations to accelerate discovery through a continuous, virtuous cycle in which AI predictions improve as new experimental results are generated, reducing the number of experimental cycles required and increasing confidence in candidate selection.

Built on Revvity Signals' deep expertise in scientific data capture, secure collaboration, and bench-scientist workflows, Signals Xynthetica delivers AI and machine learning capabilities as a service — a Models as a Service (MaaS) approach — making advanced predictive tools accessible to scientists on demand, without the need to build or maintain underlying infrastructure. The platform is designed to scale AI-augmented discovery across entire R&D organizations.

The Modern AI-Driven Design Cycle

• Artificial intelligence and machine learning are transforming the traditional Design-Make-Test-Decide workflow, with the greatest impact concentrated in the Design phase.
• Fast, data-driven in silico exploration has evolved into its own continuous cycle that guides and accelerates the broader discovery process.
• The modern design cycle begins with generative AI proposing novel candidate structures, followed by machine learning models predicting properties at scale, and concludes with virtual screening to identify the most promising candidates for experimental follow-up.
• By focusing effort on high-value designs rather than brute-force enumeration, scientists can move from ideas to synthesized compounds with greater precision and confidence.

AI-Powered In Silico Design Capabilities

• Data-guided Molecular Generation: Generative models — including large language models and diffusion models — learn from high-quality experimental data to propose new molecules that match predefined property profiles or design constraints. These models capture hidden structure–activity relationships and suggest candidates beyond classical intuition, enabling exploration of regions of chemical space that would be difficult to reach through manual design alone.
• Hybrid Property Prediction: Newly generated molecules are evaluated using a combination of machine learning and physics-based models to estimate key properties, liabilities, and developability metrics. Machine learning models are continuously updated as fresh experimental results become available, while physics-based simulations provide mechanistic insight, together delivering a more reliable and nuanced view of each candidate's potential.
• Multi-objective Optimization: Candidate designs are ranked against multiple objectives simultaneously — such as performance, manufacturability, and risk-related constraints. The system balances these often-competing criteria to identify candidates with the strongest overall trade-offs, producing a focused shortlist for experimental follow-up.

Data Fuels AI — AI Empowers Scientists

• Models embedded in scientific workflows: Predictive models are available as a service directly inside Signals One™, Signals Synergy™, and Signals ChemDraw™, ensuring that predictions are part of routine scientific decision-making and that AI-augmented design runs continuously as research progresses.
• Private fine-tuning with fresh data: As new experimental results are generated, organizations can privately fine-tune selected models on their own high-value data, sharpening prediction accuracy for their specific chemistry and biology without exposing proprietary information.
• Governed models with scalable impact: Models are managed, versioned, and governed centrally while being applied consistently across teams and projects, ensuring predictions remain transparent, auditable, and reproducible so that AI-augmented design can scale reliably across the organization.

Signals Xynthetica is currently available for early access pre-registration. Revvity Signals has a proven track record of turning complex scientific technologies into secure, easy-to-use software services, and Xynthetica extends this strength to AI-augmented design by delivering predictive models as a cloud-based service integrated directly into existing Signals platform products.