Casey P. Shannon | Search bioRxiv

bioRxiv2016

DIABLO - an integrative, multi-omics, multivariate method for multi-group classification

Systems biology approaches, leveraging multi-omics measurements, are needed to capture the complexity of biological networks while identifying the key molecular drivers of disease mechanisms. We present DIABLO, a novel integrative method to identify multi-omics biomarker panels that can discriminate between multiple phenotypic groups. In the multi-omics analyses of simulated and real-world datasets, DIABLO resulted in superior biological enrichment compared to other integrative methods, and achieved comparable predictive performance with existing multi-step classification schemes. DIABLO is a versatile approach that will benefit a diverse range of research areas, where multiple high dimensional datasets are available for the same set of specimens. DIABLO is implemented along with tools for model selection, and validation, as well as graphical outputs to assist in the interpretation of these integrative analyses (http://mixomics.org/).

Bioinformatics↗

bioRxiv2016

A novel approach to identifying marker genes and estimating the cellular composition of whole blood from gene expression profiles

Measuring genome-wide changes in transcript abundance in circulating peripheral whole blood cells is a useful way to study disease pathobiology and may help elucidate biomarkers and molecular mechanisms of disease. The sensitivity and interpretability of analyses carried out in this complex tissue, however, are significantly affected by its dynamic heterogeneity. It is therefore desirable to quantify this heterogeneity, either to account for it or to better model interactions that may be present between the abundance of certain transcripts, some cell types and the indication under study. Accurate enumeration of the many component cell types that make up peripheral whole blood can be costly, however, and may further complicate the sample collection process. Many approaches have been developed to infer the composition of a sample from high-dimensional transcriptomic and, more recently, epigenetic data. These approaches rely on the availability of isolated expression profiles for the cell types to be enumerated. These profiles are platform-specific, suitable datasets are rare, and generating them is expensive. No such dataset exists on the Affymetrix Gene ST platform. We present a freely-available, and open source, multi-response Gaussian model capable of accurately predicting the composition of peripheral whole blood samples from Affymetrix Gene ST expression profiles. This model outperforms other current methods when applied to Gene ST data and could potentially be used to enrich the >10,000 Affymetrix Gene ST blood gene expression profiles currently available on GEO.\n\nKey PointsO_LIWe introduce a model that accurately predicts the composition of blood from Affymetrix Gene ST gene expression profiles.\nC_LIO_LIThis model outperforms existing methods when applied to Affymetrix Gene ST expression profiles from blood.\nC_LI

Bioinformatics↗