Covering: 1999 to 2014 Type 1, α/β hydrolase-like thioesterase (TE) domains are essential offloading enzymes, releasing covalently bound products from fatty acid, polyketide, and non-ribosomal peptide biosynthetic complexes. The release step can occur by attack of an exogenous nucleophile effecting hydrolysis or transesterification or by an intramolecular O-, N-, or C-nucleophile, effecting macrolactonization, macrolactamization or Claisen-like condensation of the product. Thus in addition to ensuring turnover of the pathway, TEs provide access to increased chemical diversity. We review the diversity, structure, and mechanism of PKS and NRPS TEs and discuss recent works that highlight the role of TEs as potential arbitrators in offloading. In particular, we examine cases where TEs act as logic gates that ask a particular question about the substrate and use this information to determine the substrate's fate. As the TE mechanism occurs via two steps, we analyze both the loading and release steps independently as logic gates. The use of logic gates provides an important perspective when evaluating the evolution of TEs within a pathway, as well as highlighting work towards the goal of predicting TE function in unknown and engineered pathways.

ClusterMine360 (http://www.clustermine360.ca/) is a database of microbial polyketide and non-ribosomal peptide gene clusters. It takes advantage of crowd-sourcing by allowing members of the community to make contributions while automation is used to help achieve high data consistency and quality. The database currently has >200 gene clusters from >185 compound families. It also features a unique sequence repository containing >10 000 polyketide synthase/non-ribosomal peptide synthetase domains. The sequences are filterable and downloadable as individual or multiple sequence FASTA files. We are confident that this database will be a useful resource for members of the polyketide synthases/non-ribosomal peptide synthetases research community, enabling them to keep up with the growing number of sequenced gene clusters and rapidly mine these clusters for functional information.

Microbial natural products have played a key role in the development of clinical agents in nearly all therapeutic areas. Recent advances in genome sequencing have revealed that there is an incredible wealth of new polyketide and non-ribosomal peptide natural product diversity to be mined from genetic data. The diversity and complexity of polyketide and non-ribosomal peptide biosynthesis has required the development of unique bioinformatics tools to identify, annotate, and predict the structures of these natural products from their biosynthetic gene clusters. This review highlights and evaluates web-based bioinformatics tools currently available to the natural product community for genome mining to discover new polyketides and non-ribosomal peptides.