We are pleased to announce the publication of this third Database Issue of Plant and Cell Physiology (PCP). It contains four new databases and seven updated databases (Tables 1, 2). Our aim with this issue is to provide a forum for discussion of bioinformatics research, in particular the development and maintenance of the infrastructure of web databases for plant science (Matsuoka and Yano 2010). The databases described in this issue cover a broad range of omics topics. The genome and transcriptome databases permit management of the flood of data from recent high-throughput sequencers, and have been rapidly extended to apply to non-model plants. On the other hand, metabolome and phenome data still require databases and web tools to store, annotate and compare the data. In the following paragraphs, we briefly introduce the 11 databases in this issue and broadly describe their functions. Databases of physiological data are particularly interesting from a PCP standpoint. The Chloroplast Function Database (Myouga et al. 2013) offers phenotype data obtained from 2,495 Arabidopsis mutants. The data include visually identifiable mutant phenotypes and plastid ultrastructures. UniVIO (Sakakibara et al. 2013) combines hormone concentration (hormonome) and transcriptome data for 14 organ parts of rice plants at the reproductive stage, and of seedling shoots of three gibberellin signaling mutants. RAP-DB has played a fundamental role in rice research (Sakai et al. 2013). Gene structures and annotations have been updated in it ever since 2005. The database now incorporates literature-based annotations, and RNA-Seq data and single nucleotide polymorphism (SNPs) can be queried using a customized browser. Similar genomeand expressed sequence tag (EST)-based databases have been developed for orchid (Su et al. 2013, Tsai et al. 2013) and radish (Shen et al. 2013). Although the amount of information available is rather limited in these non-model plants, these databases generate a new impetus for elucidating the biological mechanisms in these species and their relatives. In addition to databases based purely on experimental data, this issue also includes databases populated by computational predictions. To unravel the stress regulome, STIFDB2 (Naika et al. 2013) combines stress-responsive transcription factors with binding sites that have been predicted by Hidden Markov models for Arabidopsis and rice. ERISdb (Szcześniak et al. 2013) predicts cis-regulatory motifs for splicing events in eight plant species. KNApSAcK-3D (Nakamura et al. 2013) and PRIMe (Sakurai et al. 2013) both provide new tools for metabolome research. KNApSAcK-3D predicts the three-dimensional structure of all the metabolic compounds in the KNApSAcK database. PRIMe provides a series of tools