The importance of curated informatics resources: data and tools – the gateway to knowledge in biology
One of the major challenges in modern systems biology is the integration of diverse high and low throughput measurements on biological systems and deciphering biological mechanisms and pathways. My laboratory has been engaged in multi-omics multi-scale data integration for over a decade. For nearly a decade, we worked on integrating diverse data from measurements on macrophages and the progress in this area is reported across several publications. First, we used time series measurements of phosphoproteins in ligand treated macrophages along with time series cytokine readouts to understand the cellular signaling processes associated with macrophage cells.
Transcripts form one of several components, albeit an important one, in a cell. My laboratory had engaged for a few years already in characterizing a few other components, microRNA, lincRNA and phosphoproteins using existing methods. In addition, we participated in a large glue grant effort on lipidomics, which witnessed the development of the state of the art mass spectrometric methods for quantitative measurements of lipid analytes. I was the Director of the Bioinformatics and Systems Biology Cores of this grant and we had a ten-year investigation which led to arguably the best lipid resource in the world, “the Lipidomics Gateway” (http://www.lipidmaps.org). Our efforts in this project resulted in numerous peer-reviewed publications.
Prior to this latest review period we had built a large number of resources (used widely by the scientific community) including the Biology Workbench, the Signaling Gateway (originally, a collaboration with Nature), the Functional Genomics Workbench (based on novel algorithms for gene expression analysis, functional annotation and pathway building), and a number of boutique data resources. Our theme in building networks from data involves identifying pathway modules either dictated by comparative and evolutionary biology consideration or derived by applying various statistical learning strategies augmented by applying biological constraints. We have built numerous methodologies over the past 10 years towards this end. Recently, we have developed algorithms to use various graph theory methods to identify modules as well as reduce complex graphical networks into unique assembly of modules. Visualization and interactive analysis of pathways is a complex problem and we have now built and released a novel pathway editing tool that combines most recent computer science methods along with innovative biological tools.
I was recently awarded a $6 million grant from NIH Common Funds to establish the national metabolomics resource. My laboratory will coordinate activities of 6 other resource centers and numerous individual investigator awards. (http://www.metabolomicsworkbench.org)