Coevolution-based prediction of protein–protein interactions in polyketide biosynthetic assembly lines

Abstract

Motivation Polyketide synthases are enzymes that generate diverse molecules of great pharmaceutical importance, including a range of clinically used antimicrobials and antitumor agents. Many polyketides are synthesized by cis-AT modular polyketide synthases (PKSs), which are organized in assembly lines, in which multiple enzymes line up in a specific order. This order is defined by specific protein-protein interactions. The unique modular structure and catalyzing mechanism of these assembly lines makes their products predictable and also spurred combinatorial biosynthesis studies to produce novel polyketides using synthetic biology. However, predicting the interactions of PKSs, and thereby inferring the order of their assembly line, is still challenging, especially for cases in which this order is not reflected by the ordering of the PKS-encoding genes in the genome. Results Here, we introduce PKSpop, which uses a coevolution-based protein-protein interaction prediction algorithm to infer protein order in PKS assembly lines. Our method accurately predicts protein orders (93% accuracy). Additionally, we identify new residue pairs that are key in determining interaction specificity, and show that coevolution of N- and C-terminal docking domains of PKSs is significantly more predictive for protein-protein interactions than coevolution between ketosynthase and acyl carrier protein domains. Availability The code is available on http://www.bif.wur.nl/ (under ‘Software’) Supplementary information Supplementary data are available at Bioinformatics online.

PolyQ length co-evolution in neural proteins

“…More broadly, as suggested by the observation that AARs in general display highly complex and interrelated evolutionary dynamics ( 5 , 10 ), these findings may also apply to AARs other than polyQ (S. Vaglietti and F. Fiumara, unpublished observations). The results of our analyses fit well with the view of molecular co-evolution as a driver of molecular co-adaptation, interactivity, and, ultimately, of organismal fitness ( 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 ). Indeed, we found, first, that functionally related polyQ proteins, which display polyQ-length co-variation, also have higher degrees of physical interactivity.…”

“…Molecular co-evolution is a process thought to optimize physiological performance in pairs and networks of functionally related proteins, representing a general determinant of molecular co-adaptation and evolutionary fitness ( 17 , 18 , 19 ). In fact, coordinated evolutionary changes in different proteins facilitate the establishment or refinement of molecular interactions ( 18 , 19 , 20 ), to the point that the detection of molecular co-evolution between proteins predicts their functional and interaction dependencies ( 21 , 22 , 23 , 24 ).…”

Metabolic pathway assembly using docking domains from type I cis-AT polyketide synthases

“…However, not all recombinant PKSs could be assembled and possess biosynthetic capacity. To understand the rules that governed interactions between these DDs, researchers constructed a predictive code to determine the specificity of PKS subunit interactions by a small number of residues in the C-terminus (head, H) and N-terminus (tail, T) 34 , 41 , 70 . Phylogenetic clustering analysis based on co-evolution protein-protein interaction corresponding to structural classification suggested that there are mainly three mutually incompatible classes ( H1a–T1a , H1b–T1b , and H2–T2 ), and a single PKS complex might contain the DDs from multiple classes 34 , 41 , 43 , 70 .…”

The year 2020 in natural product bioinformatics: an overview of the latest tools and databases

“…The new version 4 of PRISM 12 has improved chemical structure prediction capabilities, which made it possible to train machinelearning models to predict the biological activity of BGC products based on these structure predictions. Two new algorithms, DDAP 36 and PKSpop, 37 provide improved prediction of docking domain interactions between polyketide synthases, which determine the order of these enzymes in the assembly lines, and thus also the order of the incorporated monomers in their nal products. To go from monomers towards nal products, another group published a machine-learning method that predicts macrocyclization patterns for both polyketides and nonribosomal peptides.…”