Dotmatics

Luma Platform Overview

Luma is a versatile multimodal scientific intelligence platform that supports diverse scientific disciplines and techniques by enabling low-code app creation with configurable workflows, data integration, AI/ML capabilities, and secure governance, facilitating scientific discovery and R&D across fields such as drug discovery, materials science, and biotechnology through adaptive, API-driven infrastructure and tailored user experiences.

Multimodal Scientific Intelligence Platform

Luma is a multimodal scientific intelligence platform designed to support a wide range of scientific and IT disciplines, R&D modalities, and scientific techniques. It enables work in areas such as chemistry, biology, small molecules, vaccines, antibodies, materials science, data science, cell and gene therapy, enzymes, computer science, IT administration, RNA therapeutics, and chemicals & materials.

Supported Scientific Techniques and Data Modalities

  • Assay development
  • Protein characterization
  • Sequence analysis & cloning
  • Flow cytometry
  • Structured & unstructured files (numeric, text, images)
  • Experimental metadata
  • SMILES/InChI, FASTA/FASTQ/PDB, SBML/FCS2
  • Multiomic & clinical data

Configurable Low-Code Luma Apps

Luma allows users to build their own low-code scientific apps with configurable capabilities, including:

  • Open API (REST, GraphQL, JDBC)
  • Material & ontology management
  • Data management & processing
  • Adaptive workflows
  • AI/ML applications
  • Instrument integration
  • FAIR data principles
  • Data lakehouse storage
  • Tailored user experiences
  • Governance & security

Luma Composability

Luma combines technology and science in a low-code approach, offering:

  • App authoring
  • Scientific tools
  • Adaptive workflow technology
  • Step types (Make, Test, Decide)
  • Data integration & processing
  • Configuration via APIs
  • Infrastructure with Kubernetes
  • Experience builder for tailored user interfaces

Application Across Scientific Discovery

Luma apps drive outcomes in various domains:

  • Therapeutics and drug discovery
  • Chemicals, materials, and agritech
  • Protein production and cell line development
  • Assay and library development
  • Target identification and catalysis
  • Battery technology, industrial enzyme development, food additives
  • Agricultural product discovery and genetic engineering
  • Biopolymers, biofuels, and public health surveillance

Metadata Abstractions and SDLC

Luma apps use configurable metadata abstractions to blend code-based and no-code configurations, supporting coherent software development lifecycle (SDLC) characteristics. This includes design-time configuration and runtime end-user experiences.

App Extensions and Ecosystem

Specialized scientific software enhances Luma by participating in data flows and workflows. Each piece of software has a corresponding Luma app, serving as an independent ingress interface. Apps are versioned packages that encapsulate configuration by use case and can contain configuration records from any category.

Example Apps

  • Geneious App
  • BioGlyph App
  • Registration App
  • Prism App
  • Protein Metrics App
  • OMIQ/FCS Express App

Apps can declare dependencies to flow data between them. Configuration records are stored in PostgreSQL and include data integration, data management, material & ontology management, adaptive workflows, tailored user experiences, and artificial intelligence.

Scientific Software & Data Ecosystem

Luma integrates with a variety of scientific software and instruments, parsing and structuring data from instruments, files, and databases. The architecture adheres to domain-driven design practices and is implemented across 30+ Luma services.

High-Level Architecture

  • Customer network
  • Dotmatics AWS (single tenant and multitenant)
  • Frontend and shared/domain-specific Luma services
  • RESTful/HTTP ingress & egress
  • AWS API Gateway
  • Micro frontends
  • Data management (Aurora PostgreSQL Serverless)
  • Molecular design and instrument management
  • SQS queues, file parsers, directory watchers
  • Lambda functions for data processing
  • S3 storage
  • Decision support UI
  • Chemistry & biology smarts
  • Data flows for modeled and unmodeled relational tables

Plug & Play Architecture

Luma supports end-to-end workflows and enriched data acquisition using a low-code approach. It integrates with over 300 instrument categories, including:

  • Plate readers (absorbance, fluorescence, luminescence)
  • Characterization instruments (LCMS, NMR, XRD, IR-spec, Raman Spec)
  • Functional instruments (DSC, TGA, DVS, BET surface area)
  • Diagnostics (FCS, cell counters, cell analyzers)
  • Image analysis (HC imagers, microscopy)
  • Biologics (bioreactors, protein characterization)

Instrument and file-specific parser logic transforms unstructured data to semi-structured JSON, executed in efficient lambdas and stored in Databricks. Agents are remotely managed and monitored.

Data Management and Adaptive Workflows

  • Automated business logic is encapsulated, centralized, and extensible within configurable low-code data flows.
  • Transformation from semi-structured to modeled data is supported.
  • User-interactive dry lab analysis and wet lab operations are orchestrated with configurable low-code task-based workflows.
  • Container operations, expression engine, and app interoperability allow for highly specialized steps.

Material & Ontology Management

  • Design, register, and link all entities (molecular designs, wet lab designs, assay data, characterization data, endpoint data) with workflow steps.
  • Dual lineage: data & process.

Tailored User Experiences

  • Configurable user experiences centralize decision-making with traditionally disparate data across instruments, analysis types, file types, and software.

Scientific Smarts

  • Multimodal capabilities for small molecules, peptides (HELM), proteins, and structures (glyphs).
  • Native cheminformatics and bioinformatics intelligence.
  • Search by substructure and BLAST homology.
  • Integrated plasmid & linear views, alignment views (nucleotide, protein, whole genome).

Accelerated Insights and Outcomes

  • Contextualized decision-making at every level.
  • Visualization and analysis of well-correlated complex data.
  • Rapid identification of patterns and insights across data sets.
  • Digital data capture powers adaptive simulation and AI.
  • Enrichment of existing systems via open APIs.

Luma is designed to fit the way science is done, adapting to workflow and data variability, breaking down silos, increasing business agility, and delivering faster time to market, lower costs, and new discoveries.