Led by Prof Claude Chelala.
- Identify barriers preventing data being used to their maximal potential and develop bioinformatics methods to address these.
- Work closely with basic and clinical scientists to provide the bioinformatics backbone to support cutting edge research and translate data into diagnostics and therapeutics.
- Breast Cancer Now Tissue Bank bioinformatics portal (BCNTBbp)
- The IT and bioinformatics platform for the Pancreatic Cancer Research Fund Tissue Bank (PCRFTB)
BCNTBbp – The Breast Cancer Now Tissue Bank bioinformatics portal (BCNTBbp) is a major component of the Breast Cancer Now Tissue Bank (BCNTB). It provides researchers with a sophisticated literature-based data-mining tool and the analytical functionalities to conduct bioinformatics analyses on publicly available datasets. In the future, we will expand to data generated using BCNTB tissues.
PCRFTB – Supported by the Pancreatic Cancer Research fund, the PCRFTB aims to create an auditable clinical annotations database. To achieve this, a unique web-based platform capable of storing and linking clinical and pathological information from patients with different types of pancreatic cancer to donor data has been developed.
PED – The Pancreatic Expression Database (PED) is the only device currently available for mining of pancreatic cancer literature data. It brings together the largest collection of multi-dimensional pancreatic data from the literature. PED allows the user to ask specific questions on the observed levels of deregulation among a broad range of specimen/experimental types including healthy/patient tissue and body fluid specimens, cell lines and murine models as well as related treatments/drugs data.
SEARCHBreast – SEARCHBreast is a secure online database of tissues, resources and information derived from animal models of breast cancer. It encourages sharing between researchers to improve efficiency in animal research, with the aim of reducing the use of animal models overall.
SNPnexus – SNPnexus is a web-based tool that provides an aggregate set of functional annotations for genomic variation data by characterising related consequences at the transcriptome/proteome levels with in-depth analysis of potential deleterious effects, inferring physical and cytogenetic mapping, reporting related HapMap data, finding overlaps with potential regulatory, structural as well as conserved elements and retrieving links with previously reported genetic disease studies.
O-miner – The O-miner analytical suite provides the research community with the means to conduct transcriptomic and genomic analyses on both publicly available and in-house data.
We have expertise on the analysis of microarray and next-generation sequencing (NGS) data. Our computational work has been primarily focused on solid (pancreas, breast and prostate) and haematological (Follicular Lymphoma and Acute Myeloid Leukaemia) malignancies studied within BCI.
Gene expression prognosis signature
Using publicly available mRNA abundance datasets, we performed a large retrospective meta-analysis on pancreatic ductal adenocarcinoma (PDAC) patients to discover prognostic gene signatures. We identified a 36-gene signature able to prognosticate PDAC independent of patient cohort and microarray platforms. Further work will focus on understanding the functional roles, downstream events and interactions of the signature genes.
We conducted a comprehensive analysis of pancreatic cancer-expression space by integrating data on normal pancreas, normal-adjacent pancreas, pancreatic cancer, pancreatic cancer cell lines and xenografts. All findings are available from PED for independent interrogation.
We are using an integrated computational and molecular approach to examine whole-genome gene expression in breast cancer epithelial field cancerisation and related concomitant changes in the microenvironment. Our aim is to assess the clinical value of field markers in predicting the risk of cancer initiation, development, prognosis and recurrence.
Global analysis of human and murine breast cancer
To complement the SEARCHBreast online resource, a large-scale comparison of transcriptomics data from mouse model, patient tissues and breast cancer cell lines is in progress, to help us understand breast cancer models better.
Analytical modules for BCNTBbp
To complement the BCNTBbp we are developing an integrated analytical layer within the web interface for tailored analysis of key datasets, allowing users to simply and efficiently pose questions not addressed in the original publications. Four analysis categories can be performed: molecular classification; tumour purity estimates; gene expression; and survival.
We are conducting the largest multi-cohort analysis of prostate cancer mRNA abundance profiles to date using data derived from various microarray and NGS platforms. By developing an analytical framework able to detect aberrations in expression across different platforms and prostate cancer stages, we are aiming to reconstruct the natural history of the disease.
We are actively working with Prof Jude Fitzgibbon’s group to elucidate the evolution patterns driving cancer initiation and progression of follicular lymphoma and acute myeloid leukaemia.
Tawana K et al. (2015) Disease evolution and outcomes in familial AML with germline CEBPA mutations. Blood, 126(10):1214-23. PMID: 26162409
Rosalind C et al. (2015) BCCTBbp: The Breast Cancer Campaign Tissue Bank bioinformatics portal. Nucleic Acids Research. 43(Database issue):D831-6. PMID: 25332396
Speirs V et al. (2015) Animal research: share surplus animal tissue. Nature, 522(7555):156 PMID: 26062499
Haider S et al. (2014) A multi-gene signature predicts outcome in patients with pancreatic ductal adenocarcinoma. Genome Medicine, 3;6(12). PMID: 25587357
Okosun J et al. (2014) Integrated genomic analysis identifies recurrent mutations and evolution patterns driving the initiation and progression of follicular lymphoma. Nature Genetics, 46(2):176-81. PMID: 24362818
Ullah D et al. (2014) The Pancreatic Expression Database: Recent Extensions and Updates. Nucleic Acid Res, 1,42(1):D944-9. PMID: 24163255
We are expanding our team and research activities and currently seek to recruit four highly motivated and talented people working on computational (dry-lab) and experimental (wet-lab) biology.
For potential collaborations, please contact Prof. Claude Chelala (email@example.com) with a brief description of your project.
If you require bioinformatics expertise in your grant application, please get in touch when you are developing your proposal so that we can advise you on the level of support and time required for the bioinformatics.
The Bioinformatics Unit is funded by Breast Cancer Now, Cancer Research UK, the Pancreatic Cancer Research Fund, the Engineering and Physical Sciences Research Council, The National Centre for the Replacement, Refinement and Reduction of Animals in Research, and Barts and The London Charity.