LIPEA: Lipid Pathway Enrichment Analysis

Aldo Acevedo, Claudio Durán, Sara Ciucci, Mathias J. Gerl, Carlo Vittorio CannistracibioRxiv 2018 · DOI

Abstract

Analyzing associations among multiple omic variables to infer mechanisms that meaningfully link them is a crucial step in systems biology. Gene Set Enrichment Analysis (GSEA) was conceived to pursue this aim in computational genomics, unveiling significant pathways associated to certain gene signatures under investigation. Lipidomics is a rapidly growing omic field, and absolute quantification of lipid abundance by shotgun mass spectrometry is generating high-throughput datasets that depict lipid metabolism in a plethora of conditions and organisms. In addition, high-throughput lipidomics represents a new important ally to develop personalized medicine approaches, investigate the causes and predict effective biomarkers in metabolic diseases, and not only. Here, we present Lipid Pathway Enrichment Analysis (LIPEA), a web-tool for over-representation analysis of lipid signatures and detection of the biological pathways in which they are enriched. LIPEA is a new valid resource for biologists and physicians to mine pathways significantly associated to a set of lipids, helping them to discover whether common and collective mechanisms are hidden behind those lipids. LIPEA was extensively tested and we provide two examples where our system gave successfully results related with Major Depression Disease (MDD) and insulin resistance.

Introduction

In the era of "omics," lipidomics has emerged as a critical field for understanding metabolic health and disease. Advances in shotgun mass spectrometry allow researchers to quantify thousands of lipids, creating complex "lipid signatures" associated with specific physiological states. However, translating these lists of lipids into biological meaning remains a significant bottleneck. Unlike genomics, which benefits from established tools like Gene Set Enrichment Analysis (GSEA), lipidomics has historically relied on manual, bias-prone literature reviews to map lipid changes to metabolic pathways. LIPEA (Lipid Pathway Enrichment Analysis) addresses this by providing an automated, statistically rigorous web-tool for over-representation analysis (ORA) of lipidomic data.

Methods

LIPEA is designed with a modern Model-View-Controller (MVC) architecture to ensure scalability and cross-platform compatibility. The system integrates several critical components to streamline the transition from raw data to biological insight.

Architecture and ID Mapping

A primary challenge in lipidomics is the lack of standardized nomenclature. LIPEA overcomes this by implementing a robust mapping system that translates identifiers from major databases—including Swiss Lipids, Lipid Maps, ChEBI, and HMDB—into standardized KEGG (Kyoto Encyclopedia of Genes and Genomes) IDs. This "universal translator" functionality is essential for interoperability across different research tools and databases.

LIPEA System Architecture: The MVC framework integrates external databases like KEGG and Lipid Maps with a PHP/Python backend.

Statistical Framework

The core of LIPEA's analysis is the Fisher Exact Test, based on the hypergeometric distribution, which determines if a set of lipids is significantly over-represented in a specific pathway compared to what would be expected by chance. To maintain statistical integrity and mitigate the "multiple testing" problem, the tool provides Benjamini-Hochberg and Bonferroni-Holm p-value corrections. The backend utilizes parallel computing via Python scripts to ensure rapid processing of high-throughput datasets.

Results and Discussion

The utility of LIPEA was validated through two primary case studies involving human and animal models.

Case Study: Major Depressive Disorder (MDD)

Using a 6-lipid biomarker panel identified in previous studies, LIPEA automatically identified glycerophospholipid metabolism (KEGG: hsa00564) as the sole significant pathway (corrected p-value < 0.05). This result matches the findings of manual expert reviews but provides a layer of statistical validation that was previously missing.

Case Study: Insulin Resistance

In a dataset of 23 Triacylglycerides (TAGs) from Rattus norvegicus, LIPEA successfully pinpointed the insulin resistance pathway (KEGG: rno04931). This aligns with the established biological link between triglyceride accumulation in skeletal muscle and impaired insulin sensitivity.

LIPEA Analysis Results Table: A sample output showing identified pathways, lipid counts, and statistical significance metrics.

Usability and Future Directions

LIPEA is a free, open-source platform that requires no programming expertise, making it highly accessible to the broader biological and clinical community. While currently limited by the class-level granularity of many public databases (e.g., KEGG), the authors intend to evolve LIPEA into a multiscale tool. As databases begin to provide more species-specific information, LIPEA will integrate this data to offer even deeper mechanistic insights.

Conclusion

LIPEA represents a significant step forward for the lipidomics community, providing a standardized and automated method for pathway enrichment. By bridging the gap between high-throughput quantification and biological interpretation, it empowers researchers to discover the collective mechanisms underlying complex lipid signatures.

Summary updated

Abstract

Abstract Motivation Analyzing associations among multiple omic variables to infer mechanisms that meaningfully link them is a crucial step in systems biology. Gene Set Enrichment Analysis (GSEA) was conceived to pursue this aim in computational genomics, unveiling significant pathways associated to certain gene signatures under investigation. Lipidomics is a rapidly growing omic field, and absolute quantification of lipid abundance by shotgun mass spectrometry is generating high-throughput datasets that depict lipid metabolism in a plethora of conditions and organisms. In addition, high-throughput lipidomics represents a new important ally to develop personalized medicine approaches, investigate the causes and predict effective biomarkers in metabolic diseases, and not only. Results Here, we present Lipid Pathway Enrichment Analysis (LIPEA), a web-tool for over-representation analysis of lipid signatures and detection of the biological pathways in which they are enriched. LIPEA is a new valid resource for biologists and physicians to mine pathways significantly associated to a set of lipids, helping them to discover whether common and collective mechanisms are hidden behind those lipids. LIPEA was extensively tested and we provide two examples where our system gave successfully results related with Major Depression Disease (MDD) and insulin re-sistance. Availability The tool is available as web platform at https://lipea.biotec.tu-dresden.de .
BibTeX
@misc{acevedo2018lipea,
  author = {Aldo Acevedo and Claudio Durán and Sara Ciucci and Mathias J. Gerl and Carlo Vittorio Cannistraci},
  title = {{LIPEA: Lipid Pathway Enrichment Analysis}},
  year = {2018},
  doi = {10.1101/274969},
  howpublished = {bioRxiv},
}
Figure