Vital-IT Computational Biology

Our knowledge management and data stewardship competencies enable us to get novel and faster insights into complex datasets. We organize, clean & control the quality of your datasets for subsequent analysis. In addition, we process, transform and align your datasets to existing standards and make them available as a database. We are specialized in handling different types of omics data (e.g. RNA-Seq, metabolomics, lipidomics, proteomics, transcriptomics, microbiomics) including sensitive data, for which we can provide bespoke solutions  (e.g. federated analysis or secure dedicated servers for remote analysis). Our specialties include: 

• FAIR Data management  
• Data and literature curation 
• Knowledge representation 

In practice: 

• Set up of a federated database for biomarker discovery in type 2 diabetes (T2D) as part of the IMI RHAPSODY consortium. In this database, 10 observational clinical cohorts with a combined total of 50K patients were standardized, harmonized and made available for statistical analysis. We developed data mining and statistical tools in an analysis package for researchers to use within the consortium. Read more and watch the video 
• Set up of a database composed of ~100 clinical trials as part of the IMI Hypo-RESOLVE consortium. The work has involved setting up ETL (extract, transform, load) pipeline for feeding trial data into a secure Oracle database and the creation of a virtual analysis environment accessible via remote desktop and secured by 2 factor authentication. 
• Set up of secure pipelines to process and analyse exome sequencing data in various cancers, in order to identify the factors responsible for resistance to immunotherapy, as part of the IMI IMMUcan project.  

More about SIB’s role in European public-private partnerships 

We propose analysis pipelines adapted to each research context, including multi-omics models and results interpretation. Our specialties include: 

• Omics analysis 
• Integrative and network analysis 
• Machine learning 

In practice: 

• Development of a method for interpreting gene expression data with metabolic models which has helped to identify key genes in obesity-related inflammation in adipose tissue (read the paper) 
• Integrative analysis of mouse and human lipidomics data which led to the discovery of novel lipid biomarker candidates for T2D (read the paper) 
• Transcript assembly and functional characterization of heart-specific lncRNAs in collaboration with the University of Lausanne Medical School (read the paper) 
• Transcriptomics data analysis from muscle biopsies in a human clinical trial investigating the effect of Urolithin A on muscle function (read the paper) 

We develop tailored web applications and software tools to present, analyze, visualize and interpret data and results. We use various standards as appropriate to ensure interoperability and reusability (e.g. W3C standards such as RDF).

In practice:

• Development, in collaboration with EMBO, of SourceData - a web-based platform for curating and annotating figures in scientific publications (read the paper).
• Development, in collaboration with the SIB Clinical Bioinformatics Group and the Geneva University Hospitals (HUG), of Oncobench ® - a software platform to help cancer diagnosis (read the news)