[Paper Review] Comprehensive Lipidomic Automation Workflow using Large Language Models

Connor Beveridge, Sanjay Iyer+6 more|PubMed|Mar 22, 2024

Metabolomics and Mass Spectrometry Studies7 citations

TL;DR

Describes CLAW, an automated lipidomics workflow with integrated MRM-based parsing, statistical analysis, and an AI-driven user interface using large language models.

ABSTRACT

Lipidomics generates large data that makes manual annotation and interpretation challenging. Lipid chemical and structural diversity with structural isomers further complicates annotation. Although, several commercial and open-source software for targeted lipid identification exists, it lacks automated method generation workflows and integration with statistical and bioinformatics tools. We have developed the Comprehensive Lipidomic Automated Workflow (CLAW) platform with integrated workflow for parsing, detailed statistical analysis and lipid annotations based on custom multiple reaction monitoring (MRM) precursor and product ion pair transitions. CLAW contains several modules including identification of carbon-carbon double bond position(s) in unsaturated lipids when combined with ozone electrospray ionization (OzESI)-MRM methodology. To demonstrate the utility of the automated workflow in CLAW, large-scale lipidomics data was collected with traditional and OzESI-MRM profiling on biological and non-biological samples. Specifically, a total of 1497 transitions organized into 10 MRM-based mass spectrometry methods were used to profile lipid droplets isolated from different brain regions of 18-24 month-old Alzheimer's disease mice and age-matched wild-type controls. Additionally, triacyclglycerols (TGs) profiles with carbon-carbon double bond specificity were generated from canola oil samples using OzESI-MRM profiling. We also developed an integrated language user interface with large language models using artificially intelligent (AI) agents that permits users to interact with the CLAW platform using a chatbot terminal to perform statistical and bioinformatic analyses. We envision CLAW pipeline to be used in high-throughput lipid structural identification tasks aiding users to generate automated lipidomics workflows ranging from data acquisition to AI agent-based bioinformatic analysis.

Motivation & Objective

• Develop an automated workflow for parsing, statistics, and lipid annotations in lipidomics data.
• Integrate multiple reaction monitoring transitions into a cohesive CLAW platform.
• Enable identification of double-bond positions in unsaturated lipids via OzESI-MRM.
• Provide an AI-driven user interface to interact with CLAW for statistical and bioinformatic analyses.
• Demonstrate the workflow on large-scale lipidomics datasets from biological and non-biological samples.

Proposed method

• Create a CLAW platform with modules for parsing and annotation of lipid species from MRM transitions.
• Incorporate OzESI-MRM methodology to identify carbon–carbon double bond positions in unsaturated lipids.
• Use 1497 transitions organized into 10 MRM methods to profile lipid droplets from mouse brains (Alzheimer’s model vs controls).
• Apply integrated statistical and bioinformatic analyses within the CLAW workflow.
• Develop an AI agent-based language UI to allow users to drive analyses via chatbot interactions.

Experimental results

Research questions

• RQ1 Can CLAW automatically parse and annotate lipids from large MRM-based datasets?
• RQ2 How effectively does OzESI-MRM contribute to double-bond position identification in unsaturated lipids within CLAW?
• RQ3 What is the scalability of CLAW for high-throughput lipidomics across biological and non-biological samples?
• RQ4 Can an AI agent-based interface enable streamlined statistical and bioinformatic analyses within lipidomics workflows?

Key findings

• CLAW handles 1497 transitions across 10 MRM methods for brain lipidomics in aged Alzheimer’s model mice.
• OzESI-MRM provides structural specificity by identifying C=C positions in unsaturated lipids.
• Canola oil TG profiling demonstrates carbon–carbon double bond specificity is achievable in CLAW workflows.
• An AI-agent powered language UI enables interaction with CLAW for statistical and bioinformatic analyses.
• The workflow supports high-throughput lipid structural identification from data acquisition to analysis.