Translating big metabolomics data into molecular knowledge
Metabolomics complements genomics, transcriptomics and proteomics by providing information on biochemical processes and revealing the contributions of non-genetic factors, such as the environment, diet and microbiome. In spatial metabolomics, the challenge is to localise hundreds of metabolites with cellular and sub-cellular resolution. An open-source bioinformatics engine Recent technological advances in high mass resolution imaging mass spectrometry have made the imaging of metabolites in biological tissue sections possible. However, a major bottleneck is the lack of bioinformatics methods for molecular interpretation of spatial metabolomics data. To address this, scientists from the EU-funded METASPACE (Bioinformatics for spatial metabolomics) project developed an open online community platform that integrated bioinformatics tools into validated workflows for clinical use. “Our ultimate goal was to generate a research ecosystem for exploiting spatial metabolomics data beneficial to both academia and industry,″ explains project coordinator Dr Theodore Alexandrov. METASPACE developed algorithms for high-throughput putative annotation of hundreds of metabolites, including a novel score for quantifying whether a metabolite is present in cells of a tissue section. For example, it can find oncometabolites in a tumour section or drugs delivered to and metabolised in the kidney cortex. These algorithms were implemented as cloud software with a user-friendly interface on the METASPACE website. The scientific team engaged the scientific community in Europe and worldwide who provided over 3 000 public datasets in the largest data sharing effort in the field of spatial metabolomics. The METASPACE team evaluated the engine in a clinical case study of oesophageal cancer as well as in drug development applications for pharmaceutical R&D. The use of cloud technologies enabled processing of more than 50 TB of raw data. The future of METASPACE METASPACE is already used by over 200 scientists from over 70 labs across the world. The project twitter account actively disseminates news and engages the community of more than 400 followers. Dr Alexandrov confesses that “this project transformed my view on modern science and its impact. It was very inspiring to see the scientific community agreeing to share their data, and thus creating a big public knowledge base of spatial metabolomes.″ Collectively, the information on the metabolites detected in tissue sections from animal models and clinical human samples lays the foundation for further research in the field of spatial metabolomics and for addressing important biomedical questions. The increasing need for metabolite annotation highlights the impact of the project especially with respect to disease-associated metabolomes. Importantly, the platform enables biologists and clinical specialists without mass spectrometry or bioinformatics experience to translate complex data into molecular knowledge. From the beginning, METASPACE was designed as an open-source and open-access project. The aim was to stimulate sustainable development in this field with a platform capable of incorporating future algorithms, software and services. Outreach efforts have led to the collaboration in the ERC Consolidator grant-funded METACELL, where the METASPACE platform will be used for detection of cellular metabolites. The engine will also be employed as a test bed in the EU ICT CloudButton project for the development of novel cloud technologies. From an industrial perspective, there is great potential for the METASPACE platform to be incorporated in the drug discovery and testing workflows of big pharma. Looking towards the future, Dr Alexandrov is confident that “METASPACE will facilitate future biomedical discoveries, support efficient therapies, and drive innovation thus creating novel R&D opportunities in Europe.″
Keywords
METASPACE, metabolite, bioinformatics, platform, spatial metabolomics, cloud, open science