IDSL.GOA: Gene Ontology Analysis for Interpreting Metabolomic datasets

Abstract

Biological interpretation of metabolomics datasets often ends at a pathway analysis step to find the over-represented metabolic pathways in the list of statistically significant metabolites. However, the map definitions and the metabolite coverage vary among different curated databases, leading to inaccurate and contradicting interpretations. For the lists of gene, transcripts and proteins, Gene Ontology (GO) terms over-presentation analysis has become a standardized approach for the biological interpretation. GO terms are not limited to predefined pathways but can also include relevant metabolic processes that are not included in pathway databases. Despite the several advantages of GO terms over traditional pathway maps, GO analysis has not been achieved for metabolomics datasets. To overcome this, we present a new knowledgebase and the online tool, Gene Ontology Analysis by the Integrated Data Science Laboratory for Metabolomics and Exposomics (IDSL.GOA) to conduct GO over-representation analysis for a metabolite list. The IDSL.GOA knowledgebase covers 2,324 metabolic GO terms and associated 2,818 genes, 22,264 transcripts, 20,158 proteins, 1,482 EC annotations, 2,430 reactions and 2,212 metabolites. IDSL.GOA analysis of a case study of older vs young female brain cortex metabolome highlighted over 250 GO terms being significantly overrepresented (FDR <0.05). The analysis suggested that in the older female brain cortex region, nucleotide salvage processes are severely affected. On contrast, for the same metabolite list, MetaboAnalyst and Reactome Pathway Analysis suggested less than 5 pathways at FDR <0.05, and none of them were related to nucleotide salvage pathways. We showed how IDSL.GOA identified key and relevant GO metabolic processes that were not mentioned by a commonly used pathway analysis approach. Overall, we suggest that metabolomics researchers should not limit the interpretation of metabolite lists to only pathway maps and can also leverage GO terms as well. IDSL.GOA provides a powerful tool for this purpose, allowing for a more comprehensive and accurate analysis of metabolite pathway data. IDSL.GOA tool can be accessed at https://goa.idsl.me/

bootGSEA: a bootstrap and rank aggregation pipeline for multi-study and multi-omics enrichment analyses

“…In particular, a GO term or pathway is only highly ranked by the integrated score if there is evidence for the importance of a term or pathway from different omics levels. We demonstrated the use of our approach in an example with transcriptomics and proteomics data, but it could be extended by GSEA from other omics domains, such as metabolomics (Mahajan et al, 2024). The rank aggregation step also supports the idea of research synthesis, that is, integrating findings from different studies or data sources to obtain a higher level of scientific evidence.…”

Metabolomic and Proteomic Analysis of ApoE4-Carrying H4 Neuroglioma Cells in Alzheimer’s Disease Using OrbiSIMS and LC-MS/MS

“…We built up metabolite data sets that were significantly different between the ApoE4 KI group and the control group from OrbiSIMS (35) and LC-MS/MS (40), respectively. Out of this list, 15 (OrbiSIMS) and 28 (LC-MS/MS) metabolites had KEGG identifiers available and were used as input for IDSL.GOA analysis . The GO analysis for OrbiSIMS metabolomics suggested a total of 17 GO processes that were changed, such as the cysteine, sulfur amino acid catabolic processes (Table S17), and tRNA aminoacylation for mitochondrial protein that were significantly affected by ApoE4.…”

“…However, the biological functions of metabolites and metabolic pathways are not comprehensively covered and can vary across different databases, which may lead to misunderstandings and poor biological interpretations. Mahajan et al 34 reported an online GO tool for metabolomic data sets that can identify important GO metabolic processes not covered in existing pathway databases. Therefore, we applied IDSL.GOA to our metabolomic data sets to gain deeper biological insights.…”

“…Out of this list, 15 (OrbiSIMS) and 28 (LC-MS/MS) metabolites had KEGG identifiers available and were used as input for IDSL.GOA analysis. 34 The GO analysis for OrbiSIMS metabolomics suggested a total of 17 GO processes that were changed, such as the cysteine, sulfur amino acid catabolic processes ( Table S17 ), and tRNA aminoacylation for mitochondrial protein that were significantly affected by ApoE4. For LC-MS/MS, 120 GO biological processes were affected by ApoE4, which involved pyridine-containing compound process, cysteine metabolic process, cellular amino acid biosynthesis process, etc.…”

“…Out of this list, 15 (OrbiSIMS) and 28 (LC-MS/MS) metabolites had KEGG identifiers available and were used as input for IDSL.GOA analysis. 34 The GO analysis for OrbiSIMS metabolomics suggested a total of 17 GO processes that , along with the proteins involved in their corresponding Gene Ontology (GO) biological processes. Log ratios between the ApoE4-carried group and the control group for each protein were mapped to the nodes using a blue−white−red gradient.…”