Deep learning models are dominating performance across a wide variety of tasks. From protein folding to computer vision to voice recognition, deep learning is changing the way we interact with data. The field of natural products, and more specifically genomic mining, has been slow to adapt to these new technological innovations. As we are in the midst of a data explosion, it is not for lack of training data. Instead, it is due to the lack of a blueprint demonstrating how to correctly integrate these models to maximise performance and inference. During my PhD, I showcase the use of large language models across a variety of data domains to improve common workflows in the field of natural product drug discovery. I improved natural product scaffold comparison by representing molecules as sentences. I developed a series of deep learning models to replace archaic technologies and create a more scalable genomic mining pipeline decreasing running times by 8X. I integrated deep learning-based genomic and enzymatic inference into legacy tooling to improve the quality of short-read assemblies. I also demonstrate how intelligent querying of multi-omic datasets can be used to facilitate the gene signature prediction of encoded microbial metabolites. The models and workflows I developed are wide in scope with the hopes of blueprinting how these industry standard tools can be applied across the entirety of natural product drug discovery. / Thesis / Doctor of Philosophy (PhD)

This project focuses on the use of sponge genetic transcripts in the form of expressed sequence tags (ESTs) readily available in Genbank to search for novel genes using bioinformatics analysis tools. Marine sponge species are known to house a diversity of marine microbes and are known as the ‘living fossils’ of the animal kingdom because of the large number of ancient genes they house. Genomic mining can be a useful tool in discovering these orthologous genes. This study utilized the techniques of genomic mining of 11 previously described sponge species transcripts. The results of this study provide a better understanding of the genomic structure of the organisms studied by creating a more detailed genetic map and examining a specific environmental snapshot of the genes in each sponge. Novel methods for dissecting beneficial information from large scale data sets available in genomic libraries utilizing bioinformatics search tool MGRAST were examined. The results of this study indicate that sponges house numerous genes that are likely to be evolutionary predecessors of genes in higher eukaryotes. Support was also given to the notion that microbial communities play a role in metabolic pathways of sponges.