Figure 2: The computationally directed genomic enzyme-mining pipeline. | Nature Communications

Figure 2: The computationally directed genomic enzyme-mining pipeline.

From: Integrative genomic mining for enzyme function to enable engineering of a non-natural biosynthetic pathway

Figure 2

GEOs were identified based on amino-acid sequence homology to KIVD. Bioinformatics filters were used to identify a set of amino-acid sequences from genomic databases that are predicted to be KIVD homologues and likely to be decarboxylases. Once identified, homology models were built to obtain a predicted ternary structure of each GEO. Ligand docking and design simulations were subsequently run in the presence of our target ligand (for example, C8) to evaluate the potential protein–ligand interface energy. A phylogenetic tree for the 239 GEOs is depicted with a bar chart above each sequence. The bar height indicates the predicted protein–ligand interface energy; the higher the bar the lower the energy. Bar height is scaled linearly relative to the lowest protein–ligand interface energy. Ten GEOs (brown) were selected for experimental characterization. The pairwise sequence identity of all 239 GEOs to each other has a mode of 20%, indicating the high level of sequence diversity within this set of sequences.

Back to article page