We develop and validate methods for organizing, summarizing and visualizing complex biological data for systems biology through the integration of bioinformatics and biostatistics.
Data sets collected for pan-cancer cell lines (mutation, methylation, copy number aberration (CNA), cancer type, gene-expression, proteomics and IC50 (drug response)) were topologically ordered. There is no direct relation between mutation, methylation, CNA, cancer type with IC50; all this information is channeled through the gene-expression. Proteomics does have a separate link with IC50 due to PTMs. Find out more here.
Masoumeh Alinaghi, Johan Westerhuis and Age Smilde developed a new data analysis method for the analysis of multiple sets of data with an underlying experiment design. The method (PE-ASCA) is able to apply ASCA models on common and distinct information from multiple data sets. PE-ASCA was applied to real metabolomics data obtained from NMR analysis of two different brains tissues (hypothalamus and midbrain) from piglets in an intervention study. The paper was highlighted on the cover of the January 2020 issue of Metabolomics.