Proteomics & Metabolomics

Proteomics Bioinformatics

The goal of proteomics research in general is to identify and quantify proteins on a defined biochemical state. This field heavily relies on computer software and computational power. The proteome informatics workflows at the FGCZ have continuously been growing combining a wide variety of open source and commercial code. The workflows are serving more than 200 scientists from University of Zurich and ETH Zurich, covering a data size of more than 20 Tera Bytes, using a cluster infrastructure with more than 300 compute nodes.

The Functional Genomics Center supports a wide range of proteomics workflows, including the traditional gel-based proteomic analysis (1-D and 2-D gels), gel-free liquid chromatography proteomic analysis (also known as shotgun proteomics), up to the analysis of protein interactions.

For the more established workflows we rely on bioinformatics software solutions from commercial vendors, such as ProteomeWeaver for the analysis of 2-D gels, Matrix Science Mascot-Server for peptide and protein identification and quantification, ProteomeSoftware Scaffold for results analysis and validation.

For large scale high throughput projects a SEQUEST pipeline with cluster support and validation with the Transproteomic Pipeline (from the Institute for Systems Biology, Seattle) is available.

Data management and annotation of all processes is ensured by the B-Fabric LIMS system. Here, internal and external customized software solutions for diverse problems are integrated for further usage. Packages like BLAST, MatrixScience Mascot Distiller, and several open source and collaboration based in-house software tools can be triggered by the users.

We also encourage and support data upload into public repositories, like PRIDE, using controlled vocabularies for annotation.

Contact

Christian Panse, Jonas Grossmann, Simon Barkow-Oesterreicher (Office: Y32H88a)

Metabolomics Bioinformatics

FGCZ offers computational facilities and training to manage and to analyse aquired metabolomics data. For the time being, metabolomics data analysis is based mainly on third party open source or commercial software. Complementing solutions and work flow automation are being developped constantly and according to actual needs.

Data Management

Data management builds on raw data file naming conventions and sample annotation which is mandatory in the moment if a metabolomics measurement is set up on one of dedicated instruments (LC-MS or ESI-FTMS systems). Raw data shall be reviewed and transfered manually to a dedicated storage volume at the FGCZ intranet by the responsible user. Transfered data will then be processed in an automated fashion with the result that raw data as well as derived data is secured and mapped in the FGCZ data management system B-Fabric.

Data Analysis

Currently we support a basic data analysis work flow for metabolomics data. It handles LC-MS (or GC-MS) data (time - m/z - intensity) representing at least two experimental conditions (e.g. controls and perturbed, before and after, time 0 and time > 0) with at least two replicates for each condition. The workflow is designed to align identical metabolites and to detect changes between the conditions. The workflow includes the following steps:

- Conversion of raw data into NetCDF or m/z xml format
- Alignement of MS traces and peak detection in the time domain
- Quantification of aligned peaks and calculation of fold change between conditions
- Listing of detected peaks, with quantitative information and possible assignements

Further support may include metabolite identification based on isotope and fragmentation patterns, pathway mapping and the visualization of metabolite-metabolite as well as metabolome-proteome-genome interactions.

Support Modes

Projects are supported in User lab and Service mode. On case by case base dedicated teaching or the development of dedicated solutions may be agreed.

Contact

Endre Laczko (Metabolomics User lab)