E Pluribus Unum: Understanding and Influencing Pluripotency & Reprogramming by Integrative Bioinformatics

点击翻译按钮获取中文摘要

Following up on work accomplished in the past 20 months since SPP funding commenced, we intend to complete machine learning of pluripotency, now focusing on data mining & integration, based on our PluriNetWork, and associated ExprEssence software that suggests mechanisms involved in pluripotency, by comparing high-throughput datasets. We will expand the PluriNetWork of pluripotency genes/proteins in mouse, and establish it for human (& other species), focusing on transcriptional regulation, signaling & epigenetics. Understanding pluripotency & reprogramming is then based on integrating differential highthroughput datasets into our networks and identifying putative mechanisms, comparing a) different species, b) different data types (protein, RNA, epigenetics, …), and c) different interventions & time series, different states of pluripotency (ICM, Epiblast, iPS, …) & different reprogramming approaches (iPS, SCNT, fusion). We dissect pluripotency & reprogramming in an analysis across all dimensions, using the networks to integrate the data, and to propose intervention schemes (integrating small molecules). The differential analyses by ExprEssence are developed further to cope with these new tasks, incl. components for generating movies, highlighting the startup & shutdown of mechanisms in a network along all dimensions. ExprEssence highlighting will thus identify ensembles of subnetworks to be subjected to detailed modeling (Theis), it will support Boiani, Greber, Meisterernst, Hiiragi, Schroeder, Tanaka, Besser, F.-J. Müller, A. Müller & Zenke, and it is also the future core of the Bioinformatics Helpdesk for the SPP.