Gene Expression

Fios Genomics can analyse gene expression profiling data from all commercial and custom microarray platforms and exon arrays. We offer computational innovative solutions to help you overcome challenges in genomic microarray analysis and achieve biologically insightful results rapidly. From data generation, quality and statistical filtering to network and mechanistic pathway biology — we have optimised analysis solutions to suit all your needs and streamline workflow for predictive accuracy and cost-effective report outputs.

View the Fios Genomics workflow here.

• Experimental design consultation

  Careful experimental design is the first key step in getting the most from microarray studies. We offer a Consultation Service to assist you in the choice of microarray platform, advice on wet lab. service providers and options for balanced and statistically sound experimental design.

• Differential gene expression

  Data QC and normalisation

  QC identifies arrays that are in some way defective, either as a result of a poor labelling, a poor hybridisation, or other properties that confer outlier status relative to the other arrays. The parameters assessed depend to a degree on the array platform being considered, but the primary goal is to ensure that the data being submitted to subsequent analyses are of the highest possible quality.

  Statistical analysis

  Knowing which test to apply to which experimental paradigm requires an expert knowledge of statistics, enabling the optimal identification of statistically differentially expressed genes. We can perform a range of robust statistical tests from the simple to the more advanced depending on the experimental design. The output of statistical analyses consists of differential gene expression measurements and their corresponding statistical significance taking into account biological variation and number of samples used.

• Network analysis

  Using novel network visualisation tools it is possible to construct large 3D networks from microarray data, whereby co-expressed genes form highly connected cliques of inter-connected nodes. This new and powerful approach to data analysis is particularly suited to the exploration of large and complex datasets, in which technical or biological variation may give rise to high dimensional structure within the data. This platform also supports the integration of data from other sources. A network graph component prior to clustering is shown on the right.

• Gene clustering

  Using network-based clustering algorithms it is possible to cluster genes based on their expression profile, identifying cohorts of co-expressed genes. Co-expressed genes often represent functional units of biology, share common regulatory motifs and provide the key to understanding the core biological facets of an experiment. Clustering of the above network graph component is shown on the right. The clusters within the single network graph are shown in different colours.

• Functional and pathway analysis

  In order to help interpretation, lists of differentially expressed genes or gene clusters can be mined to assess the enrichment of functionally related genes. This can include Gene Ontology terms, biological pathways and user-defined sets of genes with known attributes. Lists can also be explored for known relationships e.g. protein-protein interactions between members.

• Results interpretation and visualisation

  All of the above analyses are summarised to interpret the data in the context of known relationships between genes to provide hypotheses on functionality or biological response. Examples of novel relationships between genes, if present, can be identified.

  All of the above analyses will be performed in a short and defined period of time agreed with the client. The output from the analyses is provided to the client as a comprehensive report in a "publication-ready" format.

• Uploading your microarray data to a public repository

  Upon receiving sufficient data and information to fulfil MIAME criteria, we will submit your data to GEO.