This document discusses large-scale integration of data and text in bioinformatics. It describes using text mining on millions of abstracts and articles to extract information on biological entities and their associations in order to build networks of proteins, genes, diseases and small molecules. This information is integrated with experimental data and computational predictions into web-centric databases and resources that can help researchers by saving them time over manually reviewing the literature. Visualization tools are also provided to project network data onto tissue and subcellular localization information extracted from text.