This thesis reports the investigations of class I lyase-like enzymes with particular focus on aiding discovery of family members, and variants thereof, for biotechnological applications, such as biocatalytic synthesis of value-added compounds.Chapter 2 details computational investigations of class I lyase-like enzymes based on sequence and structural data. Using an initial set of structurally- and biochemically-characterised class I lyase like enzymes, patterns and relationships were identified and used to annotate publically-available sequences. This allowed the discovery of potential enzyme-coding genes for use in areas of biotechnology, e.g. as biotherapeutics for the treatment of cancer or as biocatalysts for the production of valuable unnatural amino acids. The search also aided elucidation of putative biosynthetic pathways, including one for a narrow spectrum antibiotic, and highlighted possible mechanisms of functional evolution within the family. In chapter 3 the characterisation and engineering of the bacterial ammonia lyase EncP for the production of (S)-β-amino acids is reported. This enzyme, although previously reported in the literature, had ever been investigated in a biocatalytic context. Creation of a biotransformation method allowed the broad substrate scope and clear enantiopreference of the enzyme to be uncovered. By combining electronic effects of substrates with structural inference, it was possible to create enzyme variants with shifted regioselectivity, including EncP-R299K - a biocatalyst catalysing the (S)-β-selective amination of a range of acrylic acids. This result is complementary to previous work as the (S)-β-products were not previously obtainable using already characterised ammonia lyase biocatalysts.Chapter 4 is about the use of another biocatalyst, AvPAL, to perform preparative scale synthesis of (S)-alpha-amino acids. Upon investigation of the substrate scope of this enzyme, imperfect enantio- and regioselectivity were uncovered. Further investigation of the product mixtures revealed that the enzyme had unreported mutase-like side activity, pointing to evolutionary mechanisms of functionalisation, as relating to chapter 2. Unfortunately engineering efforts to augment these activities were relatively unsuccessful. By choosing optimal substrates and reaction conditions, a biotransformation method was developed, allowing industrially relevant space time yields (up to 60 g L-1 d-1) to give crude isolated amino acids in sufficient purity.Chapter 5 provides further details on exact computational and experimental methods used throughout the investigations.