| Lignin is a recalcitrant complex aromatic polymer found in the cell walls of plants and some algae and is involved in structural support, pathogen defense, and water transduction. The scientific and economic importance of lignin can be attributed to its abundance, useful chemical properties, and connection to plants and their evolution. The advent of Next Generation Sequencing (NGS) technologies has provided researchers with new tools to study lignin metabolism via bioinformatics. Most of the currently available bioinformatics tools rely on sequence similarity for the identification of new sequences. However, the volume of data generated by NGS is fast outpacing our ability to biochemically study enzymes. This is cumbersome for sequence similarity-based methods as they require experimentally validated homologues to function well. Alternatively, Artificial Intelligence (AI) or machine learning-based techniques are an elegant solution to these problems as they can accurately classify sequences irrespective of homology. In this study, we developed a highly accurate computational system capable of identifying lignin metabolism related enzymes based on the reactions they catalyze. Sequence similarity techniques can also cloud functional understanding of enzyme subtypes. Laccases, an economically important enzyme linked to degradation and synthesis of lignin, are a good example of this. They are involved in a wide range of metabolic functions and have broad but specific substrate ranges. These properties are not reflected by current multiple sequence alignment-based classification systems. To address this problem, we first created an unsupervised learning system capable of making descriptor-based classification systems and tested it using the enzyme class 'Laccases'. Secondly, a bioinformatics pipeline was developed to identify and classify all possible lignin-related enzymes. Finally, the computational algorithms developed from this study were implemented as web-based bioinformatics tool(s), available freely at http://bioinfo.okstate.edu/ to aid the researchers working in the area of lignin. These tools should also have high applications to the biofuel industry. |
