Bioinformatics Seminar
Time:
Venue: Na
12 November 2019
NaStable genes as anchor points in transcriptomic analysis
Dharmesh BhuvaWEHI Bioinformatics
Transcriptomic signatures are useful in understanding the molecular phenotypes of cells ; tissues ; and patient samples ; for example ; they have been successfully applied to stratify breast cancer patients into molecular subtypes. In most cases ; gene expression signatures are developed using whole-transcriptome scale measurements. This means that methods for matching signatures to samples typically require samples to be measured on the same scale. The need for relatively large amounts of starting material ; and sequencing cost for whole-transcriptome measurement limits clinical applications ; and accordingly thousands of existing gene signatures are unexplored in a clinical context.
Genes within a signature carry most of the information about the molecular phenotype they represent ; so an efficient assay and scoring method would quantify the abundance of these genes with few additional measurements needed. We have modified our method ; singscore ; to quantify and summarise relative expression levels of signature genes from individual samples through the inclusion of additional "stably-expressed genes". We identified genes with stable expression across different abundances and with a preserved relative order across large numbers (thousands) of samples to allow signature scoring ; as well as to support general data normalisation. We show that signature scores computed from whole-transcriptome data are comparable to those calculated using only values for signature genes and our panel of stable genes. Additionally ; we show that the list of stable genes we propose are more stable in cancer and normal tissue datasets than previously proposed stable genes. The new approach to gene signature scoring opens up the potential to develop panel-type tests for gene expression signatures that can support clinical translation of the powerful insights contributed by transcriptomic studies in cancer.;;;;