Bioinformatics

Bioinformatic analysis

Our team of bioinformaticians has extensive experience in analysing datasets generated by high-throughput technologies and supports research projects that employ a range of experimental techniques, including: 

• RNA-seq for differential gene expression between sample groups, e.g., treated vs. untreated cell lines, and single cell RNA-seq for identification and characterisation of sub-populations of cells 

• Whole genome, exome, and amplicon sequencing to explore variation in cancer genomes, identifying mutations, copy number aberrations and genomic rearrangements 

• Spatial transcriptomics to determine where different genes are expressed in a tissue sample 

• Nanopore long-read sequencing 

• Quantitative proteomics including tandem mass tag mass spectrometry to detect changes in protein abundance following treatment or a perturbation 

Statistical design and analysis

We support a wide range of statistical designs to improve signal-to-noise ratio, reproducibility and external validity of preclinical experiments that aim to advance our understanding of the causes of cancers. We determine appropriate statistical methods and apply these to a wide variety of complex data types. Some examples of the type of work we carry out include:

• Design and analysis of preclinical in vivo experiments, e.g. antitumour activity experiments
• Regression models (e.g. ANOVA, generalized linear models) with fixed and random effects applied to cancer progression and treatment response
• Parametric and distribution-free models for survival analysis
• Rule-based and model-based designs for dose-finding and dose-response studies

Software development

We have developed several software packages for analysing and visualizing the data sets we work with. These include analysis pipelines for processing data at scale, interactive web applications, web services and the relational databases that these access, and bespoke tools written in various programming languages, mainly R, Python and Java.

We also provide expert advice to other groups in the institute on best practices for developing robust software solutions. Many of our software packages and tools are available on GitHub.

Genomic data processing

The Bioinformatics Core is actively involved in processing increasingly large volumes of genomic sequencing data and develops pipelines to run these analyses efficiently on the CRUK CI high-performance compute cluster.

Working closely with the Genomics Core, we play an active role in the smooth running of the genomic sequencing operation and are responsible for developing and maintaining the LIMS deployment and automated workflows for data processing and QC. This includes the large-scale management and delivery of sequence data to hundreds of users.

Training

An important role of the Bioinformatics Core is to provide training to CRUK CI scientists and, working with the University of Cambridge Bioinformatics Training Facility, we offer a range of classroom-based training courses with an emphasis on hands-on, practical-based learning.

We teach courses on experimental design, statistical modelling and analysis, analysis of bulk and single-cell RNA-Seq data, and data science using the R programming language.

More details on our training courses can be found here.