We discuss an applied study in cancer genomics that integrates data and inferences from laboratory experiments on cancer cell lines with observational data from human breast cancers. The biological focus is on improving understanding of transcriptional responses of cells to changes in acidity in the cellular environment, and our integrative analysis aims to connect experimentally defined biomarkers of such responses to clinical outcomes in breast cancer. The analysis is a case study that exemplifies a general strategy for this kind of integration: connecting patterns of biological response linked to specific experimental interventions into observational studies where such responses may be evidenced via variation in gene expression across samples, with potential to define biomarkers of clinically relevant physiological states and outcomes. Statistical methods use Bayesian analysis with sparse latent factor regression models to identify, explore and relate signatures of aggregate gene expression changes between laboratory and observational studies. Identifying potential clinically useful prognostic factors can help to direct future laboratory studies as well as generate potential for therapeutic advances.

Data and input/output files from the analyses in this case study are available here.

• In Vitro signature analysis: Text files containing the data, BFRM input input and parameter setting files and all BFRM output files from the analysis of the neutralisation/lactic acidosis intervention data. Further information on running BFRM and using output files is available at the BFRM software page.
• In Vivo factor analysis details: Text files containing the data, BFRM input input and parameter setting files and all BFRM output files from the sparse latent factor analysis of the Miller breast tumor data.
• Survival analysis details: Text files containing the data, SSS survival analysis input and parameter setting files and all output files for the survival analysis of the Miller breast tumor data, using latent factors, signatures, and clinical variables as possible covariates. Further information on running SSS and using output files is available at the SSS software page.
• Matlab Scripts: Matlab scripts for generating the figures present in the manuscript.
• Validation Data: Matlab workspaces containing the Pawitan and Sotiriou breast tumor data sets, along with the imputed latent factor scores associated with these samples.